© 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. 37273
    Prev
    Next
    INDEX
    Explanations

    references to contemporary themes in various contexts

    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
    288
    +0.15
    0.7%
    339
    +0.13
    0.6%
    623
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    339
    +0.15
    0.01
    623
    +0.13
    0.01
    34
    +0.10
    0.01
    Negative Logits
    phy
    -0.97
    edin
    -0.94
    unal
    -0.84
    aye
    -0.81
    bows
    -0.79
    urat
    -0.78
    alties
    -0.77
    lua
    -0.76
    rations
    -0.76
    boa
    -0.75
    POSITIVE LOGITS
     sensibilities
    0.90
     trends
    0.81
     Perspect
    0.79
     perspectives
    0.78
     incarnation
    0.78
     versions
    0.77
     Antiqu
    0.77
     relevance
    0.76
     realities
    0.76
     art
    0.76
    Activations Density 0.010%

    No Known Activations