© 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. 26588
    Prev
    Next
    INDEX
    Explanations

    references to official records and documents

    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
    447
    +0.15
    0.7%
    642
    +0.11
    0.5%
    140
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    402
    +0.15
    0.02
    140
    +0.11
    0.02
    386
    +0.10
    0.02
    Negative Logits
    Interstitial
    -0.98
    Stars
    -0.74
    ategory
    -0.72
    nen
    -0.72
    âĶģ
    -0.70
    ichita
    -0.68
    cffff
    -0.68
     4090
    -0.66
    nir
    -0.65
    awar
    -0.65
    POSITIVE LOGITS
     documents
    1.14
     detailing
    1.08
     pertaining
    1.03
     outlining
    0.98
     declass
    0.97
     documenting
    0.96
     retrieved
    0.95
     evid
    0.94
     relating
    0.94
    arians
    0.94
    Activations Density 0.022%

    No Known Activations