© 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. 0-RES-JB
    11. 4046
    Prev
    Next
    INDEX
    Explanations

    exclamations or statements emphasizing a strong auditory experience

    oai_token-act-pair · gpt-3.5-turbo

    references to loudness or expressions of things being loud

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ahime
    -0.72
    hips
    -0.68
     Ster
    -0.67
    ago
    -0.66
    PG
    -0.66
    ania
    -0.66
    azo
    -0.66
    ski
    -0.66
     Scholarship
    -0.66
    olo
    -0.64
    POSITIVE LOGITS
     loud
    3.91
     louder
    2.74
     Loud
    2.14
     loudly
    2.09
     noisy
    1.89
     noise
    1.58
     aloud
    1.56
     loudspe
    1.49
     audible
    1.49
     quieter
    1.44
    Activations Density 0.019%

    No Known Activations