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. 8-RES-JB
    6. 7893
    Prev
    Next
    INDEX
    Explanations

    words related to fitting, suitability, or compatibility

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate suitability or compatibility

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    cember
    -0.74
    bies
    -0.67
    ials
    -0.66
    yard
    -0.66
    wcs
    -0.65
     Appears
    -0.65
     presided
    -0.63
     tending
    -0.63
    ery
    -0.63
    inent
    -0.60
    POSITIVE LOGITS
     snug
    1.17
     into
    0.91
    gerald
    0.87
     onto
    0.84
     squarely
    0.83
     precon
    0.82
     comfortably
    0.80
     neatly
    0.78
     seamlessly
    0.77
    kat
    0.75
    Activations Density 0.036%

    No Known Activations