© 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. 11-RES-JB
    11. 17049
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "well" followed by a descriptor in the text

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate knowledge or competency

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Pione
    -0.68
    omore
    -0.66
    ggle
    -0.66
    rely
    -0.64
     Trafford
    -0.63
    iolet
    -0.61
    itton
    -0.61
     Samp
    -0.58
    ulia
    -0.58
    committee
    -0.55
    POSITIVE LOGITS
    spring
    1.21
     behaved
    1.11
    esley
    0.99
     known
    0.96
     deserved
    0.95
    -
    0.94
    known
    0.92
     vers
    0.91
     respected
    0.91
    rounded
    0.91
    Activations Density 0.031%

    No Known Activations