Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 415
    Prev
    Next
    INDEX
    Explanations

    dates and schedules within a timeline structure

    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
    640
    +0.20
    0.9%
    481
    +0.18
    0.9%
    43
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.20
    0.04
    6
    +0.18
    0.04
    176
    +0.10
    0.02
    Negative Logits
     outwe
    -0.77
     minded
    -0.72
     behav
    -0.72
    matically
    -0.69
    beh
    -0.66
    ube
    -0.66
     infer
    -0.65
     someday
    -0.65
    etheless
    -0.64
     favoured
    -0.64
    POSITIVE LOGITS
     Finish
    0.82
     Recap
    0.78
     Begins
    0.76
     Earthquake
    0.76
     Appears
    0.72
     Ferry
    0.72
     Released
    0.71
    noon
    0.69
    Cele
    0.69
     Doors
    0.68
    Activations Density 0.096%

    No Known Activations