© 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. 4-RES-JB
    11. 18512
    Prev
    Next
    INDEX
    Explanations

    phrases related to size or importance

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "biggest."

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

    No Comments

    Negative Logits
    schild
    -0.78
    ences
    -0.77
    isson
    -0.76
    jad
    -0.75
    heid
    -0.75
    ttes
    -0.73
    aven
    -0.70
    anism
    -0.70
    endants
    -0.70
    arrell
    -0.69
    POSITIVE LOGITS
    gest
    1.00
     offender
    0.91
     culprit
    0.87
     bang
    0.84
     obstacle
    0.83
     hurdle
    0.77
     thing
    0.77
     difference
    0.76
     loser
    0.76
     proponent
    0.76
    Activations Density 0.019%

    No Known Activations