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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 24164
    Prev
    Next
    INDEX
    Explanations

    terms related to medical conditions

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "oc."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ĪĴ
    -0.89
    ħĭ
    -0.73
     Verse
    -0.68
     differe
    -0.67
     Furious
    -0.67
     Trin
    -0.63
    +++
    -0.63
     provocative
    -0.63
     wording
    -0.63
     hope
    -0.62
    POSITIVE LOGITS
    entric
    1.39
    abulary
    1.19
    hester
    1.13
    onut
    1.12
    ombat
    1.09
    urrencies
    1.06
    lean
    1.06
    orpor
    1.06
    ontent
    1.05
    ourse
    1.05
    Activations Density 0.012%

    No Known Activations