© 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 3072
    9. 8-RES_FS3072-JB
    11. 3041
    Prev
    Next
    INDEX
    Explanations

    references to academic publications and studies

    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
    631
    +0.11
    0.5%
    49
    +0.11
    0.5%
    47
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    631
    +0.11
    0.06
    313
    +0.11
    0.06
    586
    +0.10
    0.04
    Negative Logits
    chers
    -0.76
     tatt
    -0.73
     vans
    -0.71
     checkpoints
    -0.70
    haul
    -0.68
    HUD
    -0.67
    UGH
    -0.67
     lockout
    -0.67
     raids
    -0.66
    bucks
    -0.66
    POSITIVE LOGITS
     Proceedings
    1.26
     Pediatrics
    1.23
     Biology
    1.18
     Psychiatry
    1.17
     Neuroscience
    1.15
     Methods
    1.12
    doi
    1.12
     PLoS
    1.10
     Perspect
    1.09
    ournals
    1.07
    Activations Density 0.861%

    No Known Activations