© 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. 0-RES-JB
    11. 1209
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "photo" or similar variations thereof

    oai_token-act-pair · gpt-3.5-turbo

    references to various forms of the word "photo."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     cracked
    -0.76
     cracking
    -0.69
     adj
    -0.68
     robust
    -0.67
     Cran
    -0.66
     skew
    -0.63
     diversity
    -0.63
     crack
    -0.62
     episodes
    -0.60
     Aud
    -0.60
    POSITIVE LOGITS
    oto
    4.69
    OTO
    2.30
    otos
    2.14
    otto
    1.20
    otor
    1.19
    ono
    1.17
    udo
    1.17
    odo
    1.13
    ota
    1.12
    ogo
    1.12
    Activations Density 0.018%

    No Known Activations