© 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. 3-RES-JB
    11. 11458
    Prev
    Next
    INDEX
    Explanations

    proper nouns, specifically the name "Trent" at varying activation levels

    oai_token-act-pair · gpt-3.5-turbo

    references to the character "Trent."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     curs
    -0.66
     transitional
    -0.63
     subsistence
    -0.61
    rm
    -0.60
     serious
    -0.60
     Absent
    -0.59
    BIP
    -0.59
    shapeshifter
    -0.58
    lly
    -0.58
    overs
    -0.58
    POSITIVE LOGITS
     Rez
    1.29
     Trent
    1.05
    anus
    0.84
    seys
    0.83
    aments
    0.80
    ulla
    0.79
     Baal
    0.79
    ón
    0.79
    orial
    0.79
    onite
    0.78
    Activations Density 0.005%

    No Known Activations