Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 7596
    Prev
    Next
    INDEX
    Explanations

    alphanumeric codes, discount instructions, technical terms, and salutations

    oai_token-act-pair · gpt-3.5-turbo

    mentions of promotional offers and benefits

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    assetsadobe
    -0.84
    wav
    -0.83
    Ñı
    -0.81
    ÑĮ
    -0.74
    rast
    -0.74
    å§«
    -0.74
    STD
    -0.73
     Palest
    -0.71
    Ry
    -0.69
     Frie
    -0.68
    POSITIVE LOGITS
     add
    1.86
    add
    1.82
     Add
    1.80
     ADD
    1.76
    Add
    1.76
     added
    1.60
     adds
    1.54
    ADD
    1.50
     addon
    1.49
     Adding
    1.48
    Activations Density 0.212%

    No Known Activations