© 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. 37399
    Prev
    Next
    INDEX
    Explanations

    vocative phrases or expressions of exclamations

    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
    373
    +0.14
    0.6%
    537
    +0.09
    0.4%
    314
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    41
    +0.14
    0.03
    373
    +0.09
    -0.01
    284
    +0.09
    0.03
    Negative Logits
    FORMATION
    -0.60
     Divinity
    -0.59
     Curiosity
    -0.57
     FTC
    -0.57
    âĢ¢âĢ¢
    -0.56
     invention
    -0.55
    URES
    -0.55
     ted
    -0.54
     fulfillment
    -0.54
     satisfied
    -0.54
    POSITIVE LOGITS
    pillar
    1.05
    apest
    1.00
    wana
    1.00
    lehem
    0.93
    zinski
    0.85
    hammad
    0.83
    ombo
    0.83
    abase
    0.82
    reau
    0.82
    levard
    0.81
    Activations Density 0.032%

    No Known Activations