© 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. 4-RES-JB
    11. 15335
    Prev
    Next
    INDEX
    Explanations

    technical jargon and isolated keywords, especially those related to computer science and data analysis

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Further," indicating an emphasis on additional information or continuation of a topic

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

    No Comments

    Negative Logits
    liest
    -0.77
    ldom
    -0.73
    boy
    -0.70
    cius
    -0.66
    istine
    -0.64
    RN
    -0.64
    ARS
    -0.64
    esp
    -0.63
     jaws
    -0.63
    lua
    -0.63
    POSITIVE LOGITS
     ado
    0.95
    ance
    0.88
     Reading
    0.83
     refinement
    0.81
    backer
    0.80
     clarification
    0.75
     Further
    0.74
    Discussion
    0.74
    rawdownloadcloneembedreportprint
    0.73
    ãĤ´ãĥ³
    0.72
    Activations Density 0.020%

    No Known Activations