© 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. 3-RES-JB
    11. 7295
    Prev
    Next
    INDEX
    Explanations

    terms related to vehicle exhaust systems

    oai_token-act-pair · gpt-3.5-turbo

    terms related to exhaust systems

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Pound
    -0.83
    HCR
    -0.79
    AFTA
    -0.69
     Acqu
    -0.65
    Edited
    -0.65
    llan
    -0.65
     Lans
    -0.62
     prisoner
    -0.61
     Ibid
    -0.61
    EEEE
    -0.61
    POSITIVE LOGITS
    ãĥ¼ãĥĨãĤ£
    1.06
    ively
    0.99
    haust
    0.99
     manifold
    0.96
     fumes
    0.95
     exhaust
    0.95
    ives
    0.93
     emission
    0.85
    ptions
    0.83
     manif
    0.80
    Activations Density 0.019%

    No Known Activations