© 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. 0-RES-JB
    11. 22308
    Prev
    Next
    INDEX
    Explanations

    words starting with 'unc'

    oai_token-act-pair · gpt-3.5-turbo

    terms related to being uncategorized or unclassifiable

    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
     Towers
    -0.75
    brance
    -0.74
    heed
    -0.71
    ento
    -0.70
    rine
    -0.69
    rooms
    -0.69
    phe
    -0.68
    meet
    -0.68
    sbm
    -0.68
    orsche
    -0.67
    POSITIVE LOGITS
     unc
    3.47
    unc
    1.57
     unf
    1.54
     unm
    1.42
     Unc
    1.38
     unsc
    1.38
     unch
    1.37
     unb
    1.33
     unt
    1.33
     unw
    1.29
    Activations Density 0.014%

    No Known Activations