Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 46183
    Prev
    Next
    INDEX
    Explanations

    instances of the word "the" 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
    393
    +0.19
    0.8%
    385
    +0.16
    0.7%
    362
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    385
    +0.19
    0.03
    362
    +0.16
    0.03
    393
    +0.12
    0.02
    Negative Logits
    ãĥĦ
    -0.86
    leeve
    -0.74
    <?
    -0.72
    ivas
    -0.70
    ãĥīãĥ©
    -0.70
    è¦
    -0.69
    among
    -0.69
     Tradable
    -0.68
    aiden
    -0.67
     besides
    -0.66
    POSITIVE LOGITS
     end
    1.09
     expiration
    1.05
     onset
    1.03
     earliest
    1.02
     arrival
    1.01
     next
    1.01
     latter
    1.00
     final
    0.99
     entirety
    0.96
     slightest
    0.96
    Activations Density 0.056%

    No Known Activations