© 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. 7-RES-JB
    11. 23439
    Prev
    Next
    INDEX
    Explanations

    time durations in the format of "X and a half" where X is a whole number

    oai_token-act-pair · gpt-3.5-turbo

    references to the measurement of time, specifically related to "a half" and similar phrases indicating duration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Reviewer
    -0.81
    ãĥ¤
    -0.79
    thumbnails
    -0.78
    Trend
    -0.70
    berus
    -0.68
    Thom
    -0.68
    antha
    -0.67
    Prev
    -0.66
    acco
    -0.65
    ukong
    -0.64
    POSITIVE LOGITS
     dozen
    0.81
     million
    0.74
     shy
    0.72
     minutes
    0.70
     ago
    0.68
    hearted
    0.66
     marathon
    0.66
     stint
    0.63
     apiece
    0.62
    quel
    0.62
    Activations Density 0.011%

    No Known Activations