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

    Neuronpedia

    APICircuit 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. 0-RES-JB
    6. 970
    Prev
    Next
    INDEX
    Explanations

    words related to audio recordings or transcripts

    oai_token-act-pair · gpt-3.5-turbo

    various forms of the word "record" and related terms

    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
    emi
    -0.74
     mir
    -0.70
    MAP
    -0.65
     chill
    -0.64
     mosqu
    -0.64
    ère
    -0.63
     movements
    -0.62
    Mot
    -0.62
    ENSE
    -0.61
    ika
    -0.60
    POSITIVE LOGITS
    ording
    3.79
    ordable
    1.32
    oyle
    1.14
     ante
    1.08
    content
    1.07
    ulin
    1.05
    ofer
    1.00
    atered
    0.96
    onse
    0.94
    luence
    0.90
    Activations Density 0.044%

    No Known Activations