© 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. 914
    Prev
    Next
    INDEX
    Explanations

    proper nouns, specifically focusing on surnames

    oai_token-act-pair · gpt-4-turbo

    terms related to specific names or persons, particularly "Townsend"

    oai_token-act-pair · gpt-3.5-turbo

    references to specific nouns or names, particularly those associated with individuals and preferences

    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
    earchers
    -0.98
     Gutenberg
    -0.83
    orically
    -0.82
    ouri
    -0.80
    ors
    -0.78
    edient
    -0.78
    matically
    -0.77
    orable
    -0.77
    videos
    -0.75
    onial
    -0.74
    POSITIVE LOGITS
    UTC
    0.93
    IELD
    0.73
    %%
    0.70
    KK
    0.66
    coat
    0.65
    iculty
    0.65
    Dial
    0.64
    fork
    0.63
    onso
    0.63
     Osw
    0.63
    Activations Density 0.022%

    No Known Activations