© 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. 7700
    Prev
    Next
    INDEX
    Explanations

    phrases related to interruptions or breaks, especially in a schedule or activity

    oai_token-act-pair · gpt-3.5-turbo

    mentions of break periods or intervals in various 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
    ifice
    -0.74
    Reviewer
    -0.74
    IFIED
    -0.73
    itatively
    -0.72
    ãĥĺãĥ©
    -0.72
    MSN
    -0.70
     obser
    -0.69
    velength
    -0.68
     Cosponsors
    -0.67
    urally
    -0.61
    POSITIVE LOGITS
    away
    1.31
    neck
    1.25
    fast
    1.07
    aways
    1.02
    downs
    0.94
    water
    0.93
    points
    0.89
    down
    0.88
    through
    0.88
    robe
    0.87
    Activations Density 0.039%

    No Known Activations