© 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. 3-RES-JB
    11. 6075
    Prev
    Next
    INDEX
    Explanations

    mentions of paper or related material

    oai_token-act-pair · gpt-3.5-turbo

    references to different types of paper

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    akening
    -0.74
    ostic
    -0.73
    CVE
    -0.71
    oise
    -0.70
    cffffcc
    -0.70
    igious
    -0.68
    alez
    -0.67
    cius
    -0.66
    ean
    -0.66
    ois
    -0.66
    POSITIVE LOGITS
    clip
    1.20
     towels
    1.09
     towel
    0.95
    clips
    0.91
    backs
    0.86
     pus
    0.86
    Paper
    0.86
    flies
    0.83
    maid
    0.81
    pee
    0.78
    Activations Density 0.018%

    No Known Activations