© 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 3072
    9. 8-RES_FS3072-JB
    11. 3022
    Prev
    Next
    INDEX
    Explanations

    instances of when someone is asked or forced to do something

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    665
    +0.14
    0.6%
    197
    +0.12
    0.5%
    742
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.14
    0.13
    70
    +0.12
    0.10
    217
    +0.10
    0.07
    Negative Logits
     Confeder
    -0.64
    ENT
    -0.60
     convergence
    -0.58
    FK
    -0.55
     formation
    -0.55
     Integration
    -0.55
    Est
    -0.54
    tein
    -0.54
     emphasis
    -0.54
    fish
    -0.53
    POSITIVE LOGITS
     aback
    1.10
     by
    1.06
    pez
    0.89
     BY
    0.84
    by
    0.83
    uled
    0.83
    otyp
    0.74
    wic
    0.74
     repeatedly
    0.73
     merciless
    0.71
    Activations Density 1.024%

    No Known Activations