© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 2-RES-JB
    11. 23335
    Prev
    Next
    INDEX
    Explanations

    words related to sulfates, sulfides, and sulfur compounds

    oai_token-act-pair · gpt-3.5-turbo

    references to sulfur and its various chemical forms or compounds

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Wired
    -0.73
     Puzzles
    -0.72
    HEAD
    -0.72
    Eva
    -0.71
     Patri
    -0.71
    GAME
    -0.69
     Cipher
    -0.68
    Constructed
    -0.68
     Dracula
    -0.68
    gerald
    -0.67
    POSITIVE LOGITS
     sulf
    1.25
     sulph
    1.15
    urous
    0.93
     oxide
    0.91
    atile
    0.89
     sulfur
    0.85
     dioxide
    0.85
    chlor
    0.83
     iod
    0.83
    uth
    0.81
    Activations Density 0.006%

    No Known Activations