© 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. 1-RES-JB
    11. 13400
    Prev
    Next
    INDEX
    Explanations

    instances of the word "standing" in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "standing"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ffic
    -0.84
     Success
    -0.70
    ilation
    -0.69
    rez
    -0.68
    vertis
    -0.67
    cision
    -0.63
    ologist
    -0.63
    metry
    -0.63
    urer
    -0.63
    escal
    -0.63
    POSITIVE LOGITS
     standing
    1.02
     upright
    0.91
    still
    0.90
    anke
    0.88
    Standing
    0.78
     detrim
    0.76
     ashore
    0.72
    hof
    0.70
     onstage
    0.70
     downstairs
    0.68
    Activations Density 0.010%

    No Known Activations