© 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. 4-RES-JB
    11. 23972
    Prev
    Next
    INDEX
    Explanations

    the word "bat" at various activation levels

    oai_token-act-pair · gpt-3.5-turbo

    the English word "bat" and its various forms or related references

    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
     marrow
    -0.64
    ŃĶ
    -0.62
    icer
    -0.60
     resil
    -0.60
    base
    -0.59
    iley
    -0.58
     Pace
    -0.57
    erve
    -0.57
    pires
    -0.57
    mble
    -0.56
    POSITIVE LOGITS
    tered
    0.95
    iation
    0.80
    iated
    0.80
    ting
    0.76
    chers
    0.76
    hetically
    0.73
    ters
    0.71
    rix
    0.70
    rogen
    0.70
    hesda
    0.70
    Activations Density 0.025%

    No Known Activations