© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 18888
    Prev
    Next
    INDEX
    Explanations

    references to specific sports events and accomplishments

    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
    245
    +0.13
    0.6%
    156
    +0.10
    0.5%
    373
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    636
    +0.13
    0.03
    245
    +0.10
    0.03
    24
    +0.10
    0.03
    Negative Logits
    ngth
    -0.69
     lapse
    -0.64
     solicit
    -0.62
    iste
    -0.61
    hur
    -0.61
    ATED
    -0.60
    Website
    -0.60
    rea
    -0.59
    zik
    -0.59
     burner
    -0.59
    POSITIVE LOGITS
     outright
    0.89
     convinc
    0.85
     lottery
    0.84
    tein
    0.82
     decisively
    0.80
     trophies
    0.79
     toss
    0.78
     hearts
    0.77
     Pulitzer
    0.76
     battles
    0.76
    Activations Density 0.117%

    No Known Activations