Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 11197
    Prev
    Next
    INDEX
    Explanations

    technical terms related to science and technology, specifically in the field of biochemistry

    oai_token-act-pair · gpt-3.5-turbo

    scientific terms and language related to biological and technological processes

    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
    ividual
    -0.78
     ItemLevel
    -0.61
    ecause
    -0.61
    olean
    -0.56
    onga
    -0.56
    ogun
    -0.54
    wark
    -0.53
    him
    -0.53
    ruary
    -0.52
    bara
    -0.52
    POSITIVE LOGITS
     antioxid
    0.87
     Seym
    0.81
    ß
    0.81
    kefeller
    0.80
     tradem
    0.79
     millenn
    0.77
    Balt
    0.76
    EStream
    0.75
    HUD
    0.75
    natureconservancy
    0.75
    Activations Density 15.004%

    No Known Activations