© 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. 18501
    Prev
    Next
    INDEX
    Explanations

    references to negative or challenging situations

    oai_token-act-pair · gpt-3.5-turbo

    instances of past actions or events described by adjectives or adverbs that imply repetition or ongoing processes

    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
    aturday
    -0.49
    orem
    -0.48
     SHARES
    -0.45
    Published
    -0.45
     (?,
    -0.44
     Blazers
    -0.42
    reads
    -0.42
    ategories
    -0.41
    undrum
    -0.41
    odcast
    -0.40
    POSITIVE LOGITS
    ).[
    0.65
    !).
    0.62
    ?).
    0.60
    ).
    0.52
    )."
    0.51
    ]."
    0.50
     thereto
    0.49
     ).
    0.47
     nonetheless
    0.47
    %).
    0.46
    Activations Density 5.179%

    No Known Activations