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. 0-RES-JB
    6. 17774
    Prev
    Next
    INDEX
    Explanations

    references to written documents, especially manuscripts

    oai_token-act-pair · gpt-3.5-turbo

    references to manuscripts and related terms in documents

    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
     Hik
    -0.67
     inst
    -0.62
    Zone
    -0.62
    occup
    -0.61
     Ley
    -0.60
    eper
    -0.60
     Duty
    -0.59
     Oxy
    -0.59
     zones
    -0.58
     amenities
    -0.58
    POSITIVE LOGITS
     manuscript
    3.73
     manuscripts
    2.91
    uscript
    2.51
     unpublished
    1.38
     parchment
    1.16
     screenplay
    1.10
     transcription
    1.09
     paperback
    1.09
     handwritten
    1.06
     dissertation
    1.02
    Activations Density 0.036%

    No Known Activations