© 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. 10146
    Prev
    Next
    INDEX
    Explanations

    phrases related to physical movement and actions

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks, particularly commas, indicating pauses or breaks in dialogue or narrative structure

    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
    reprene
    -0.71
    onym
    -0.70
    ocl
    -0.66
    olo
    -0.65
    cia
    -0.65
    mite
    -0.63
    Commerce
    -0.63
    pec
    -0.63
    uitive
    -0.63
    ongevity
    -0.63
    POSITIVE LOGITS
     albeit
    0.96
     whence
    0.92
     although
    0.90
     lest
    0.85
     namely
    0.80
     while
    0.79
     though
    0.79
     whereas
    0.79
     but
    0.78
     presumably
    0.77
    Activations Density 0.979%

    No Known Activations