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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 7-RES-JB
    6. 17471
    Prev
    Next
    INDEX
    Explanations

    various types of actions or concepts that are labeled or denoted by the term "calls"

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating someone citing or referencing something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    chant
    -0.81
    ysc
    -0.80
    earances
    -0.72
    ooter
    -0.70
    oult
    -0.70
    roleum
    -0.69
    umber
    -0.67
    eor
    -0.67
    inth
    -0.66
    psons
    -0.66
    POSITIVE LOGITS
     "
    0.78
     '
    0.73
     Fra
    0.67
     Quality
    0.66
     ``
    0.65
     "(
    0.63
     "-
    0.62
    Fra
    0.62
     "â̦
    0.62
     BS
    0.61
    Activations Density 0.042%

    No Known Activations