© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 3-RES-JB
    11. 5541
    Prev
    Next
    INDEX
    Explanations

    names or terms related to specific locations or entities, particularly those related to intelligence or military operations

    oai_token-act-pair · gpt-3.5-turbo

    references to specific groups or entities, particularly those related to terrorism or conflict

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ãĥ¥
    -0.69
    ochemical
    -0.68
     Hatch
    -0.66
    Í
    -0.66
     Marino
    -0.65
     Rus
    -0.64
     Pokémon
    -0.62
    cipl
    -0.61
    agers
    -0.61
    ott
    -0.61
    POSITIVE LOGITS
    ĪĴ
    0.94
    sf
    0.86
    ħĭ
    0.82
    ahime
    0.78
    querque
    0.76
    isites
    0.75
    ezvous
    0.75
    incerity
    0.74
     srf
    0.72
    topic
    0.72
    Activations Density 0.040%

    No Known Activations