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 24576
    5. 8-RES_FS24576-JB
    6. 11734
    Prev
    Next
    INDEX
    Explanations

    mentions of organizational boards or groups

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    373
    +0.10
    0.5%
    287
    +0.10
    0.5%
    339
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    287
    +0.10
    0.02
    404
    +0.10
    0.02
    611
    +0.09
    0.02
    Negative Logits
    aan
    -0.75
    ãĤ®
    -0.74
    ï¸
    -0.69
    ashtra
    -0.69
    vell
    -0.67
    ãĤ¶
    -0.65
    nah
    -0.64
    ãĥķ
    -0.64
    pread
    -0.62
    secure
    -0.62
    POSITIVE LOGITS
     Directors
    1.15
     directors
    1.04
     Governors
    0.97
     governors
    0.94
     trustees
    0.92
     Commissioners
    0.79
     advisors
    0.79
     Nursing
    0.75
     Inquiry
    0.74
     supervisors
    0.73
    Activations Density 0.047%

    No Known Activations