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. 12192
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "too" 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
    55
    +0.12
    0.5%
    513
    +0.12
    0.5%
    11
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.12
    0.04
    513
    +0.12
    0.04
    11
    +0.11
    0.04
    Negative Logits
    76561
    -0.88
    inarily
    -0.71
    ords
    -0.71
    issance
    -0.70
    anwhile
    -0.66
    enance
    -0.65
    craft
    -0.65
    riots
    -0.64
    here
    -0.62
    hyde
    -0.62
    POSITIVE LOGITS
     far
    0.91
     busy
    0.90
     afraid
    0.88
     risky
    0.84
     distracting
    0.84
     much
    0.83
     frail
    0.83
     tempting
    0.82
     costly
    0.81
     expensive
    0.80
    Activations Density 0.031%

    No Known Activations