© 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. 19825
    Prev
    Next
    INDEX
    Explanations

    mentions of technological details and specifications

    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
    160
    +0.16
    0.7%
    99
    +0.11
    0.5%
    513
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    99
    +0.16
    0.05
    507
    +0.11
    0.06
    693
    +0.09
    0.05
    Negative Logits
    gerald
    -0.67
    awaru
    -0.66
    rolet
    -0.64
    lehem
    -0.64
    ians
    -0.64
    etz
    -0.63
    iership
    -0.63
    ured
    -0.62
    urat
    -0.62
     "$:/
    -0.62
    POSITIVE LOGITS
    nd
    1.99
    ND
    1.13
    160
    1.03
    133
    1.01
    147
    1.01
     thirds
    0.99
    ½
    0.96
    187
    0.94
     externalToEVAOnly
    0.92
    245
    0.92
    Activations Density 0.126%

    No Known Activations