© 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. 18817
    Prev
    Next
    INDEX
    Explanations

    information related to events and activities

    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
    394
    +0.10
    0.5%
    199
    +0.09
    0.4%
    401
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    422
    +0.10
    0.02
    299
    +0.09
    0.02
    199
    +0.09
    0.02
    Negative Logits
    ieving
    -0.95
    tics
    -0.85
    rior
    -0.83
    rome
    -0.82
    ieve
    -0.77
    lete
    -0.76
    lean
    -0.75
    hovah
    -0.75
    zie
    -0.75
    zin
    -0.74
    POSITIVE LOGITS
     ceremonies
    1.06
     ceremony
    0.97
     weekend
    0.89
     Weekend
    0.84
     rites
    0.82
     Hours
    0.81
     doors
    0.80
     night
    0.78
     salv
    0.77
     kickoff
    0.77
    Activations Density 0.031%

    No Known Activations