© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 16876
    Prev
    Next
    INDEX
    Explanations

    names with the string "MAN" followed by a single-digit numeral, typically related to people or characters

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "man" in various contexts and forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     rapt
    -0.68
     altar
    -0.64
     cheek
    -0.62
    cca
    -0.60
     Scarlet
    -0.59
     gro
    -0.59
     rib
    -0.59
     cre
    -0.59
     NCT
    -0.59
     Twilight
    -0.58
    POSITIVE LOGITS
    MAN
    4.21
    man
    2.00
    MEN
    1.98
     MAN
    1.79
    Man
    1.68
    mans
    1.62
    mann
    1.42
    men
    1.40
    woman
    1.33
    BER
    1.33
    Activations Density 0.009%

    No Known Activations