© 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. 4-RES-JB
    11. 15192
    Prev
    Next
    INDEX
    Explanations

    phrases related to acting or functioning in various roles or capacities

    oai_token-act-pair · gpt-3.5-turbo

    the word "as" in various contexts, often indicating roles or functions

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

    No Comments

    Negative Logits
    comings
    -0.83
    aways
    -0.74
     Riding
    -0.72
    gem
    -0.70
    ilts
    -0.70
    coming
    -0.69
    worth
    -0.67
    brew
    -0.65
     Phant
    -0.61
    oat
    -0.61
    POSITIVE LOGITS
     opposite
    0.76
     differently
    0.74
     fid
    0.69
     behalf
    0.66
     ACTIONS
    0.66
     actionGroup
    0.66
     metic
    0.66
     liaison
    0.65
     lia
    0.65
     rall
    0.64
    Activations Density 0.124%

    No Known Activations