© 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. 37176
    Prev
    Next
    INDEX
    Explanations

    references to the "rear" aspect of vehicles and their components

    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
    596
    +0.13
    0.6%
    516
    +0.10
    0.5%
    356
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    596
    +0.13
    0.02
    435
    +0.10
    0.01
    516
    +0.10
    0.01
    Negative Logits
    ensable
    -0.90
    âĸĪâĸĪ
    -0.81
    achu
    -0.78
    anwhile
    -0.76
    etus
    -0.76
    akura
    -0.75
    etti
    -0.74
    eteria
    -0.73
    argon
    -0.72
    ergus
    -0.71
    POSITIVE LOGITS
    ward
    1.30
    wards
    0.98
     axle
    0.96
    facing
    0.92
     lobe
    0.91
     tyre
    0.88
    ranging
    0.87
    side
    0.87
    view
    0.85
     derail
    0.84
    Activations Density 0.005%

    No Known Activations