© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 8-RES-JB
    11. 12567
    Prev
    Next
    INDEX
    Explanations

    instances where the text discusses things that are not explicit or direct

    oai_token-act-pair · gpt-3.5-turbo

    negations or statements indicating something is not the case

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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

    Negative Logits
    ERG
    -0.76
    ESE
    -0.72
    RAFT
    -0.70
     Shap
    -0.70
    TIME
    -0.69
     Lack
    -0.66
    Times
    -0.65
    KEN
    -0.65
     Palestin
    -0.64
    resist
    -0.64
    POSITIVE LOGITS
     technically
    1.09
     necessarily
    1.08
     formally
    0.96
     overtly
    0.95
     directly
    0.92
     officially
    0.91
     flashy
    0.85
     explicitly
    0.85
     statistically
    0.85
     consciously
    0.85
    Activations Density 0.190%

    No Known Activations