© 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 · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 8-RES-JB
    11. 18446
    Prev
    Next
    INDEX
    Explanations

    punctuation marks marks such as periods and quotation marks

    oai_token-act-pair · gpt-3.5-turbo

    frequent use of punctuation, particularly periods

    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
     welcome
    -0.72
     Fortress
    -0.63
    onite
    -0.62
     booking
    -0.61
    ãĥ´
    -0.60
    ãĥ¼ãĥĨ
    -0.58
    Desk
    -0.58
     pill
    -0.57
     elector
    -0.57
     CCTV
    -0.56
    POSITIVE LOGITS
     ."
    1.20
     .
    1.02
     .)
    1.01
    -.
    0.94
     ..
    0.91
     ._
    0.88
    _.
    0.84
    ,.
    0.84
    shut
    0.82
    `.
    0.81
    Activations Density 0.014%

    No Known Activations