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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 2538
    Prev
    Next
    INDEX
    Explanations

    dialogue and interactions between characters

    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
    572
    +0.14
    0.7%
    762
    +0.13
    0.6%
    218
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    572
    +0.14
    0.09
    389
    +0.13
    0.07
    762
    +0.11
    0.09
    Negative Logits
    uably
    -0.80
     predominantly
    -0.78
    imeo
    -0.70
    ordes
    -0.69
    arcity
    -0.69
    ettel
    -0.68
     predomin
    -0.68
     globally
    -0.66
    clus
    -0.66
    ificantly
    -0.66
    POSITIVE LOGITS
    -"
    1.36
    —"
    1.32
    â̦"
    1.29
    ..."
    1.26
     â̦"
    1.16
    !?"
    1.12
    ?!"
    1.10
    !"
    1.06
    â̦."
    1.05
     alright
    1.04
    Activations Density 1.608%

    No Known Activations