© 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 12288
    9. 8-RES_FS12288-JB
    11. 472
    Prev
    Next
    INDEX
    Explanations

    statements contrasting different viewpoints or opinions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    616
    +0.11
    0.5%
    36
    +0.11
    0.5%
    561
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    616
    +0.11
    0.05
    376
    +0.11
    0.05
    754
    +0.10
    0.05
    Negative Logits
    RAFT
    -0.74
    run
    -0.67
    rn
    -0.65
    igan
    -0.63
    iership
    -0.63
    Untitled
    -0.62
    hip
    -0.60
    aven
    -0.59
    adin
    -0.59
    ospons
    -0.59
    POSITIVE LOGITS
     matter
    1.08
     longer
    1.07
    xious
    1.06
    etheless
    1.06
     doubt
    1.04
     kidding
    1.02
     wonder
    0.93
     sooner
    0.91
    ct
    0.90
     excuses
    0.86
    Activations Density 0.090%

    No Known Activations