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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 41339
    Prev
    Next
    INDEX
    Explanations

    hyphenated expressions or phrases that involve interruptions or breaks in thought

    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
    160
    +0.17
    0.8%
    308
    +0.14
    0.7%
    314
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    308
    +0.17
    0.04
    160
    +0.14
    0.04
    720
    +0.14
    0.03
    Negative Logits
     palm
    -0.73
     oven
    -0.73
     mail
    -0.70
     indemn
    -0.69
     mum
    -0.69
     barr
    -0.69
     pronoun
    -0.68
     handic
    -0.68
     swarm
    -0.67
     fid
    -0.67
    POSITIVE LOGITS
    perhaps
    1.41
    meaning
    1.41
    especially
    1.37
    which
    1.36
    including
    1.35
    particularly
    1.35
    that
    1.34
    they
    1.33
    albeit
    1.33
    these
    1.33
    Activations Density 0.037%

    No Known Activations