© 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. 6261
    Prev
    Next
    INDEX
    Explanations

    words referring to specific people or characters

    oai_token-act-pair · gpt-3.5-turbo

    the word "whom" in various contexts indicating relationships or references to people

    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
    pad
    -0.72
    roads
    -0.71
    termin
    -0.67
    reth
    -0.65
    construct
    -0.63
    bard
    -0.61
    ply
    -0.61
    ricks
    -0.61
    starting
    -0.60
    Tokens
    -0.60
    POSITIVE LOGITS
    soever
    2.05
     whom
    0.74
    izens
    0.72
     Allaah
    0.71
    nown
    0.71
    owship
    0.69
    selves
    0.67
    edIn
    0.66
     Pulitzer
    0.65
     dearly
    0.64
    Activations Density 0.020%

    No Known Activations