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 768
    5. 8-RES_FS768-JB
    6. 223
    Prev
    Next
    INDEX
    Explanations

    proper nouns or names starting with the letter "G"

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    698
    +0.11
    0.5%
    289
    +0.10
    0.5%
    141
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.11
    0.14
    289
    +0.10
    0.03
    314
    +0.09
    -0.00
    Negative Logits
     facult
    -0.63
    oÄŁ
    -0.62
    ĸļ
    -0.62
     derail
    -0.60
     retrospect
    -0.59
     compr
    -0.59
     concess
    -0.57
     scrut
    -0.57
     provision
    -0.55
     sight
    -0.55
    POSITIVE LOGITS
    ossip
    1.20
    iants
    1.18
    vernment
    1.13
    reetings
    1.12
    roups
    1.12
    hetto
    1.11
    raphic
    1.11
    ourmet
    1.09
    ospels
    1.09
    oliath
    1.07
    Activations Density 3.680%

    No Known Activations