© 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. 22448
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "friends" in the text

    oai_token-act-pair · gpt-3.5-turbo

    references to friends in social contexts

    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
    phas
    -0.77
    grim
    -0.72
    yss
    -0.70
    imon
    -0.66
    assion
    -0.64
    Ħ¢
    -0.63
    ©¶æ
    -0.63
    aton
    -0.62
    acted
    -0.62
     chloride
    -0.61
    POSITIVE LOGITS
    hips
    1.20
    folk
    1.05
    liest
    0.94
    liness
    0.91
    lier
    0.90
     acquaintances
    0.89
    hip
    0.87
    hall
    0.82
     friends
    0.82
    erv
    0.81
    Activations Density 0.050%

    No Known Activations