© 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 49152
    9. 8-RES_FS49152-JB
    11. 37224
    Prev
    Next
    INDEX
    Explanations

    legal terms and concepts related to regulations and rights

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    420
    +0.11
    0.5%
    567
    +0.10
    0.5%
    592
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    567
    +0.11
    0.03
    285
    +0.10
    0.02
    661
    +0.10
    0.02
    Negative Logits
     clusters
    -0.73
     flakes
    -0.70
    aptic
    -0.69
    UGE
    -0.69
    ipel
    -0.68
     Clin
    -0.68
     pessim
    -0.66
     gloom
    -0.63
    oward
    -0.63
     spike
    -0.63
    POSITIVE LOGITS
     obligation
    1.18
    verning
    1.12
     obligations
    1.12
     lawful
    1.12
     rights
    1.09
     exempt
    1.08
     lawfully
    1.07
    exempt
    1.06
     obligated
    1.05
     respons
    1.05
    Activations Density 0.421%

    No Known Activations