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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 170
    Prev
    Next
    INDEX
    Explanations

    technical specifications and requirements for gadgets or software

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    630
    +0.10
    0.5%
    426
    +0.10
    0.5%
    679
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    739
    +0.10
    0.05
    311
    +0.10
    0.01
    679
    +0.10
    0.12
    Negative Logits
    driving
    -0.70
     imagining
    -0.69
     someday
    -0.69
    20439
    -0.69
    instead
    -0.66
    ignt
    -0.65
    otin
    -0.65
    rily
    -0.63
     ecstasy
    -0.61
     somehow
    -0.61
    POSITIVE LOGITS
     Description
    1.23
    :
    1.20
     Notes
    1.18
     Edit
    1.11
    :-
    1.09
     Details
    1.07
     :
    1.03
     Comments
    1.03
    %:
    1.01
     Supported
    1.01
    Activations Density 2.727%

    No Known Activations