© 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. 7609
    Prev
    Next
    INDEX
    Explanations

    phrases related to groups or collections of people or objects

    oai_token-act-pair · gpt-3.5-turbo

    references to groups of people

    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
    Depth
    -0.77
    arium
    -0.72
    dayName
    -0.71
    lycer
    -0.69
     teasp
    -0.69
    runtime
    -0.67
     Handling
    -0.67
    Scope
    -0.67
    cape
    -0.65
    Usage
    -0.63
    POSITIVE LOGITS
     sorts
    0.95
     explorers
    0.83
     individuals
    0.83
     friends
    0.82
     kindred
    0.82
     youths
    0.81
     travellers
    0.77
     fellow
    0.76
     onlook
    0.76
     travelers
    0.76
    Activations Density 0.070%

    No Known Activations