© 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 49152
    9. 8-RES_FS49152-JB
    11. 37276
    Prev
    Next
    INDEX
    Explanations

    references to file operations and commands related to opening and closing files

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    7
    +0.11
    0.5%
    726
    +0.10
    0.5%
    246
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    726
    +0.11
    0.02
    246
    +0.10
    0.02
    59
    +0.10
    0.02
    Negative Logits
    Bey
    -0.76
     disband
    -0.65
    VICE
    -0.65
     endors
    -0.64
    iane
    -0.64
     footing
    -0.63
     Leilan
    -0.61
    ahon
    -0.60
    llor
    -0.59
    von
    -0.58
    POSITIVE LOGITS
    itors
    0.81
    ococ
    0.81
    office
    0.80
     doors
    0.77
    acles
    0.75
    GL
    0.74
    SSL
    0.72
     parentheses
    0.72
    upid
    0.72
    ject
    0.71
    Activations Density 0.023%

    No Known Activations