© 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. 1194
    Prev
    Next
    INDEX
    Explanations

    negative statements or expressions

    oai_token-act-pair · gpt-3.5-turbo

    negations and expressions of impossibility or denial

    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
     doubtless
    -0.75
    uta
    -0.72
    lesi
    -0.70
    ription
    -0.68
    rimination
    -0.67
    only
    -0.66
     merely
    -0.65
    afort
    -0.63
    pora
    -0.62
     IMAGES
    -0.61
    POSITIVE LOGITS
    Õ
    0.81
    Availability
    0.80
    íķ
    0.76
    WHERE
    0.75
    ãģĦ
    0.72
     darn
    0.72
    Reply
    0.67
    gets
    0.67
     bothered
    0.66
    aws
    0.64
    Activations Density 0.100%

    No Known Activations