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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 11-RES-JB
    11. 1298
    Prev
    Next
    INDEX
    Explanations

    words related to actions that are done the first time or at the beginning

    oai_token-act-pair · gpt-3.5-turbo

    key phrases and terms related to claims, specifically evaluating their validity and the context in which they are made

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     unfocusedRange
    -0.70
    ificantly
    -0.69
     longer
    -0.67
     more
    -0.62
    etheless
    -0.58
     better
    -0.55
     enough
    -0.55
     clearer
    -0.55
     considerably
    -0.55
     fewer
    -0.55
    POSITIVE LOGITS
    .:
    0.78
    into
    0.63
    :(
    0.62
    .�
    0.62
    .
    0.62
    %:
    0.62
     INTO
    0.62
     Into
    0.60
    !.
    0.60
    .–
    0.58
    Activations Density 0.796%

    No Known Activations