© 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 · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 19826
    Prev
    Next
    INDEX
    Explanations

    requests for feedback or input from users

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    756
    +0.10
    0.5%
    81
    +0.10
    0.4%
    131
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    307
    +0.10
    0.02
    610
    +0.10
    0.02
    81
    +0.10
    0.02
    Negative Logits
    Ö¼
    -0.66
    nesota
    -0.64
    Benz
    -0.63
    NAS
    -0.61
    aryl
    -0.60
    sticks
    -0.60
    ¶æ
    -0.60
    burning
    -0.59
     Amtrak
    -0.58
    -$
    -0.58
    POSITIVE LOGITS
    selves
    0.93
     ourselves
    0.86
     âĿ
    0.74
     sponsor
    0.73
    lder
    0.73
     discuss
    0.69
     selves
    0.69
    ndum
    0.68
     strive
    0.68
     conclud
    0.68
    Activations Density 0.032%

    No Known Activations