© 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. 13167
    Prev
    Next
    INDEX
    Explanations

    words related to information and evaluation, such as "infatuation" and "evaluation."

    oai_token-act-pair · gpt-3.5-turbo

    references to information and data analysis

    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
    ãĤ´ãĥ³
    -0.87
    ĸļ
    -0.84
    çĦ
    -0.81
     Lumpur
    -0.76
    wagen
    -0.73
    bley
    -0.72
    bye
    -0.71
    Ô
    -0.68
    steen
    -0.67
    EStream
    -0.66
    POSITIVE LOGITS
    QL
    0.82
    aukee
    0.81
    olia
    0.71
    ruct
    0.70
    arious
    0.68
    ction
    0.68
    ombat
    0.68
    andise
    0.67
    ishable
    0.66
    enza
    0.66
    Activations Density 0.045%

    No Known Activations