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 3072
    5. 8-RES_FS3072-JB
    6. 1612
    Prev
    Next
    INDEX
    Explanations

    animals and related terms, including specific types of animals and animal behaviors

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    151
    +0.13
    0.6%
    575
    +0.12
    0.5%
    181
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.13
    0.09
    288
    +0.12
    0.10
    151
    +0.10
    0.09
    Negative Logits
    oppable
    -0.78
     Wilkinson
    -0.71
    ISTER
    -0.70
    tg
    -0.65
     Seller
    -0.65
    ories
    -0.64
    stanbul
    -0.63
     Cheong
    -0.63
     Provider
    -0.63
    lain
    -0.63
    POSITIVE LOGITS
     carc
    1.27
     species
    1.03
     paws
    1.01
     hiber
    0.99
    meat
    0.96
     habitats
    0.93
     habitat
    0.93
    bite
    0.92
     claws
    0.91
     herds
    0.91
    Activations Density 0.988%

    No Known Activations