© 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. 1-RES-JB
    11. 23331
    Prev
    Next
    INDEX
    Explanations

    organizations or companies with specific initials (e.g., "HI" or "ISC") within the text

    oai_token-act-pair · gpt-3.5-turbo

    references to companies and organizations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    match
    -0.75
     Debor
    -0.74
    parts
    -0.69
    ioned
    -0.66
    lines
    -0.66
    aic
    -0.65
    paralle
    -0.64
     beauty
    -0.63
     lines
    -0.63
    Merit
    -0.63
    POSITIVE LOGITS
    HI
    1.44
    BA
    1.06
    ERAL
    1.01
    ELD
    1.00
    KE
    0.95
    WA
    0.94
    PS
    0.93
    ETA
    0.91
    HHHH
    0.91
    OSH
    0.90
    Activations Density 0.010%

    No Known Activations