© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Llama3.3-70B-IT
    3. 50-RESID-POST-GF
    4. 43427
    Prev
    Next
    INDEX
    Explanations

    the `letter` `letter` `dot` `comma` `exclamation` `question` `colon` `semicolon` `parenthesis` `hyphen` `slash` `backslash` `underscore` `ampersand` `at sign` `hash` `percent` `caret` `asterisk` `dollar sign` `pound sign` `tilde` `grave accent` `apostrophe` `quotation mark` `curly bracket` `square bracket` `angle bracket` `pipe` `plus sign` `minus sign` `equals sign` `greater than sign` `less than sign` `vertical bar` `solidus` `reverse solidus` `tilde` `grave accent` `acute accent` `circumflex accent` `diaeresis` `cedilla` `macron` `breve` `ring` `caron` `hacek` `ogham` `thorn` `eth` `alpha` `beta` `gamma` `delta` `epsilon` `zeta` `eta` `theta` `iota` `kappa` `lambda` `mu` `nu` `xi` `omicron` `pi` `rho` `sigma` `tau` `upsilon` `phi` `chi` `psi` `omega` `infinity` `partial differential` `nabla` `approximate` `identical to` `not equal to` `less than or equal to` `greater than or equal to` `element of` `not an element of` `subset of` `superset of` `subset of or equal to` `superset of or equal to` `union of` `intersection of` `empty set` `all` `exists` `there exists` `none` `and` `or` `not` `implies` `iff` `for all` `such that` `therefore` `because` `epsilon` `delta` `pi` `sigma` `mu` `omega` `lambda` `phi` `chi` `psi` `alpha` `beta` `gamma` `zeta` `eta` `theta` `iota` `kappa` `nu` `xi` `rho` `tau` `upsilon` `null` `nil` `void` `undefined` `NaN` `true` `false` `boolean` `integer` `float` `double` `string` `char` `byte` `short` `long` `object` `array` `list` `tuple` `set` `dictionary` `map` `hash map` `tree` `graph` `node` `edge` `root` `leaf` `head` `tail` `front` `back` `top` `bottom` `left` `right` `center` `middle` `start` `end` `begin` `finish` `init` `main` `run` `execute` `process` `handle` `send` `receive` `read` `write` `print` `display` `show` `hide` `open` `close` `create` `delete` `update` `save` `load` `get` `set` `add` `remove` `insert` `append` `prepend` `extend` `copy` `clone` `move` `swap` `sort` `search` `find` `filter` `map` `reduce` `fold` `aggregate` `calculate` `compute` `solve` `validate` `verify` `check` `test` `debug` `log` `trace` `warn` `error` `info` `debug` `verbose` `fatal` `assert` `throw` `catch` `try` `finally` `return` `yield` `await` `async` `go` `spawn` `channel` `select` `close` `panic` `recover` `defer` `import` `export` `package` `module` `library` `function` `method` `class` `interface` `struct` `enum` `const` `var` `let` `static` `public` `private` `protected` `abstract` `final` `override` `virtual` `readonly` `unsafe` `volatile` `transient` `synchronized` `native` `abstract` `impl` `trait` `protocol` `extension` `generics` `type` `alias` `union` `interface` `implementation` `constructor` `destructor` `instance` `static` `self` `this` `super` `base` `new` `delete` `clone` `copy` `move` `swap` `equal` `not equal` `less than` `greater than` `less than or equal` `greater than or equal` `add` `subtract` `multiply` `divide` `modulo` `bitwise and` `bitwise or` `bitwise xor` `bitwise not` `left shift` `right shift` `logical and` `logical or` `logical not` `increment` `decrement` `preincrement` `postincrement` `predecrement` `

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.3-70B-IT @ 50-resid-post-gf
    Configuration
    Goodfire/Llama-3.3-70B-Instruct-SAE-l50/Llama-3.3-70B-Instruct-SAE-l50.pt
    Prompts (Dashboard)
    10,000 prompts, 128 tokens each
    Dataset (Dashboard)
    lmsys/lmsys-chat-1m
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    íĶĪ
    -0.11
    ä¸ĬãģĮ
    -0.11
     Turnbull
    -0.09
    å¤īãĤı
    -0.09
    ÑĮÑİ
    -0.09
     Truy
    -0.09
     Congress
    -0.09
     Bols
    -0.09
     Mer
    -0.08
    ayed
    -0.08
    POSITIVE LOGITS
     sake
    0.10
    åĬª
    0.10
     basis
    0.09
     ëħĦëıĦë³Ħ
    0.09
    stride
    0.09
    롯
    0.09
    åIJ«
    0.09
     ìľĦíķľ
    0.09
    ä¼´
    0.09
    "';
    0.09
    Activations Density 0.034%

    No Known Activations