© 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. 6518
    Prev
    Next
    INDEX
    Explanations

    specific characters or symbols used as separators or identifiers in programming code

    oai_token-act-pair · gpt-3.5-turbo

    semicolons or punctuation marks indicating separate clauses or statements

    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
    urg
    -0.78
    orts
    -0.75
    ire
    -0.75
    ©¶æ¥µ
    -0.72
    unda
    -0.72
    se
    -0.72
    itates
    -0.72
    inki
    -0.71
    ħĭ
    -0.70
    itation
    -0.70
    POSITIVE LOGITS
     furthermore
    1.24
     moreover
    1.20
     alternatively
    1.14
     hence
    1.13
     meanwhile
    1.13
     consequently
    1.11
     additionally
    1.10
     namely
    1.08
     however
    1.08
     alas
    1.04
    Activations Density 0.051%

    No Known Activations