© 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 1536
    9. 8-RES_FS1536-JB
    11. 92
    Prev
    Next
    INDEX
    Explanations

    mentions of advertisements

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    +0.17
    0.8%
    230
    +0.13
    0.6%
    244
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.17
    -0.01
    230
    +0.13
    0.06
    244
    +0.10
    0.05
    Negative Logits
    ties
    -0.68
    mate
    -0.64
     retri
    -0.63
     hindsight
    -0.61
    Ļ½
    -0.60
     makeshift
    -0.60
     isolation
    -0.59
     invite
    -0.59
    graded
    -0.59
     perspect
    -0.58
    POSITIVE LOGITS
     Continue
    0.98
     Advertisement
    0.96
    advertisement
    0.85
    Skip
    0.79
    ieu
    0.73
    oth
    0.72
    usercontent
    0.72
    Advertisement
    0.71
    #$#$
    0.71
    advertising
    0.70
    Activations Density 0.830%

    No Known Activations