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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 8-RES-JB
    11. 24049
    Prev
    Next
    INDEX
    Explanations

    chemical compounds, particularly those containing the prefix "hex" or "oct"

    oai_token-act-pair · gpt-3.5-turbo

    chemical compounds, particularly hydrocarbons and octane-related terms

    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
    Interstitial
    -0.90
    utenant
    -0.80
    HCR
    -0.76
    jriwal
    -0.73
     conditioning
    -0.70
    tsky
    -0.70
    OAD
    -0.68
     Crowd
    -0.66
     Provider
    -0.66
     Preferences
    -0.65
    POSITIVE LOGITS
    agonal
    1.65
    agon
    1.33
    agons
    1.14
    agram
    1.06
    aval
    0.95
    urnal
    0.89
    angular
    0.89
    hex
    0.88
    opus
    0.88
    ogen
    0.87
    Activations Density 0.045%

    No Known Activations