© 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 12288
    9. 8-RES_FS12288-JB
    11. 518
    Prev
    Next
    INDEX
    Explanations

    phrases related to technological advancement and innovation

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    246
    +0.12
    0.5%
    407
    +0.09
    0.4%
    205
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.12
    0.10
    351
    +0.09
    0.06
    570
    +0.09
    0.02
    Negative Logits
    leted
    -0.59
    LD
    -0.56
    not
    -0.56
     Commissioners
    -0.55
     NEVER
    -0.55
     rather
    -0.55
     Seconds
    -0.54
    icho
    -0.54
    asionally
    -0.54
     denotes
    -0.53
    POSITIVE LOGITS
     anymore
    1.53
     nor
    1.17
     adequately
    1.08
     anything
    1.08
     any
    0.99
    urate
    0.99
    enough
    0.99
     satisf
    0.98
     meaningful
    0.98
     reliably
    0.96
    Activations Density 1.117%

    No Known Activations