© 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. 41746
    Prev
    Next
    INDEX
    Explanations

    words related to condescending attitudes or behaviors

    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
    311
    +0.13
    0.6%
    199
    +0.11
    0.5%
    749
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    311
    +0.13
    0.00
    749
    +0.11
    0.01
    199
    +0.09
    0.01
    Negative Logits
    ãĥĹ
    -0.81
    WER
    -0.77
    Accessory
    -0.73
     Chimera
    -0.70
    ted
    -0.69
    buster
    -0.67
    Lens
    -0.67
    ter
    -0.66
    kj
    -0.66
    Assembly
    -0.65
    POSITIVE LOGITS
    ension
    1.12
    ended
    1.00
    ending
    0.99
    ensions
    0.90
    ribed
    0.87
    itals
    0.86
    ribing
    0.85
    ursed
    0.85
    aith
    0.83
    ribes
    0.82
    Activations Density 0.006%

    No Known Activations