© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 1-RES-JB
    6. 7435
    Prev
    Next
    INDEX
    Explanations

    references to the word "Pel" and variations of it

    oai_token-act-pair · gpt-3.5-turbo

    references to the term "Pel" and its variants, indicating a focus on specific entities or themes associated with the term

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    CV
    -0.98
    displayText
    -0.79
    é¾įå¥ij士
    -0.72
     tune
    -0.72
    æĸ¹
    -0.71
    hips
    -0.70
    shapeshifter
    -0.67
     Sakuya
    -0.65
    export
    -0.64
    ãģķ
    -0.63
    POSITIVE LOGITS
    oton
    1.11
     Pel
    0.96
    onite
    0.83
    oaded
    0.80
    yss
    0.80
    oby
    0.79
     therap
    0.77
    terson
    0.77
    agic
    0.77
    lete
    0.75
    Activations Density 0.004%

    No Known Activations