© 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. 1733
    Prev
    Next
    INDEX
    Explanations

    words related to technical aspects and specifications

    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
    447
    +0.24
    1.1%
    588
    +0.10
    0.5%
    197
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    588
    +0.24
    0.06
    271
    +0.10
    0.05
    151
    +0.09
    0.06
    Negative Logits
    usalem
    -0.68
    thood
    -0.67
    xit
    -0.65
    ¬¼
    -0.58
    CHA
    -0.55
    peria
    -0.55
    arth
    -0.54
    anton
    -0.52
    aleb
    -0.52
    ĪĴ
    -0.52
    POSITIVE LOGITS
     compared
    1.09
     (~
    0.96
     indeed
    0.92
     insofar
    0.91
     considering
    0.87
     (>
    0.84
    looking
    0.81
     nowadays
    0.80
    isable
    0.78
     owing
    0.76
    Activations Density 0.279%

    No Known Activations