© 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. 3-RES-JB
    11. 1969
    Prev
    Next
    INDEX
    Explanations

    mentions of physical aggression or forceful actions

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "lash" and its variations, indicating a focus on expressions of anger or criticism

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ĨĴ
    -0.82
    ublic
    -0.70
    umer
    -0.70
     Friendly
    -0.68
    unknown
    -0.68
    ccess
    -0.67
    missions
    -0.67
    cession
    -0.65
     Private
    -0.65
     Sorceress
    -0.64
    POSITIVE LOGITS
     lash
    1.49
     lashes
    1.12
     Lash
    0.97
     lashed
    0.95
     whip
    0.82
    uate
    0.79
    oons
    0.77
     whipping
    0.76
     furnace
    0.76
    bolt
    0.74
    Activations Density 0.006%

    No Known Activations