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 3072
    5. 8-RES_FS3072-JB
    6. 356
    Prev
    Next
    INDEX
    Explanations

    sentences with a punctuation mark followed by sentences with fragmented words

    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
    480
    +0.15
    0.7%
    640
    +0.12
    0.6%
    231
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    231
    +0.15
    0.06
    388
    +0.12
    0.05
    284
    +0.10
    0.05
    Negative Logits
     exha
    -0.76
     charm
    -0.75
     underestimated
    -0.73
     preventive
    -0.71
     immersion
    -0.70
     hemor
    -0.70
     purse
    -0.70
     tradem
    -0.70
     bonded
    -0.69
     creatively
    -0.69
    POSITIVE LOGITS
    NET
    1.08
    taboola
    1.07
    htm
    0.96
    jpg
    0.95
     âĶľâĶĢâĶĢ
    0.90
    html
    0.89
    wav
    0.89
    gif
    0.88
    Net
    0.88
    COM
    0.88
    Activations Density 0.176%

    No Known Activations