© 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. 11-RES-JB
    11. 23914
    Prev
    Next
    INDEX
    Explanations

    proper nouns or names with 'K' as the prominent letter

    oai_token-act-pair · gpt-3.5-turbo

    references to specific individuals, particularly artists and performers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    catentry
    -0.70
     mantle
    -0.59
     priceless
    -0.54
     ORDER
    -0.54
     anonymously
    -0.53
     mant
    -0.52
     QC
    -0.52
     dozens
    -0.51
     thousands
    -0.51
     trope
    -0.51
    POSITIVE LOGITS
    ikuman
    1.21
     Lumpur
    1.20
    enei
    1.10
    ĸļ
    1.08
    zie
    1.04
    lyak
    1.02
    EStream
    0.99
    EStreamFrame
    0.97
    icter
    0.92
    ucha
    0.91
    Activations Density 0.194%

    No Known Activations