© 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. 2-RES-JB
    11. 24120
    Prev
    Next
    INDEX
    Explanations

    words or phrases indicated by special characters that signify a specific action or sequence

    oai_token-act-pair · gpt-3.5-turbo

    the character "¦" likely used to indicate a pause or transition in dialogue

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     cradle
    -0.76
     purse
    -0.75
     greens
    -0.74
     welf
    -0.74
     dispar
    -0.73
     symp
    -0.73
     submerged
    -0.72
     raged
    -0.72
     indemn
    -0.71
     nodd
    -0.70
    POSITIVE LOGITS
    etc
    1.01
    WHERE
    0.91
    yet
    0.89
    wait
    0.87
    they
    0.86
    >[
    0.85
    Bah
    0.83
    allows
    0.81
    why
    0.80
    where
    0.80
    Activations Density 0.027%

    No Known Activations