© 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 1536
    9. 8-RES_FS1536-JB
    11. 35
    Prev
    Next
    INDEX
    Explanations

    times and schedules mentioned in the document

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    296
    +0.12
    0.6%
    601
    +0.11
    0.5%
    96
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    296
    +0.12
    0.11
    601
    +0.11
    0.08
    96
    +0.11
    0.08
    Negative Logits
    åĬ
    -0.71
     Inher
    -0.68
     acknowled
    -0.65
     resemb
    -0.65
     wid
    -0.63
     bestowed
    -0.62
     collaborations
    -0.62
     transpl
    -0.61
    ·
    -0.59
     deriv
    -0.59
    POSITIVE LOGITS
     EDT
    1.38
     PDT
    1.32
     EST
    1.28
     PST
    1.28
     CST
    1.27
     CET
    1.22
     ET
    1.19
     curfew
    1.16
     GMT
    1.12
    noon
    1.09
    Activations Density 1.451%

    No Known Activations