© 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. 37352
    Prev
    Next
    INDEX
    Explanations

    interactions involving fear and emotional expressions

    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
    539
    +0.11
    0.5%
    323
    +0.10
    0.4%
    476
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.11
    0.02
    683
    +0.10
    0.02
    539
    +0.10
    0.01
    Negative Logits
    conservancy
    -0.89
    allas
    -0.70
     Samar
    -0.68
    Va
    -0.68
    DragonMagazine
    -0.67
     Templ
    -0.65
    ineries
    -0.64
    Benz
    -0.64
    olicy
    -0.64
     Manufact
    -0.63
    POSITIVE LOGITS
    *)
    1.44
    ]"
    1.37
    ]
    1.14
     *)
    1.10
    )"
    1.08
    ?]
    1.08
    >)
    1.07
    ]'
    1.05
    ]-
    1.03
    )/
    1.00
    Activations Density 0.052%

    No Known Activations