© 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. 9-RES-JB
    11. 3575
    Prev
    Next
    INDEX
    Explanations

    terms related to both real and imaginary aspects or elements

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions and terms indicating relationships or contrasts between concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Heads
    -0.82
    ttes
    -0.75
     Tears
    -0.73
     Cohn
    -0.73
    Stars
    -0.70
    anners
    -0.69
     guts
    -0.69
    $$$$
    -0.69
    OPLE
    -0.68
    PsyNetMessage
    -0.67
    POSITIVE LOGITS
    rogen
    0.97
    ifice
    0.95
     unprotected
    0.94
     unex
    0.92
     outgoing
    0.92
    rogens
    0.89
    organic
    0.88
     unconventional
    0.88
     unmarked
    0.87
     uncontrolled
    0.86
    Activations Density 0.183%

    No Known Activations