© 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. 0-RES-JB
    6. 2802
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Text" followed by a number indicating the strength of activation

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of 'text' or 'textbooks'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     estim
    -0.74
     agility
    -0.70
    hole
    -0.68
     endurance
    -0.67
     leagues
    -0.67
     gamble
    -0.66
    berman
    -0.66
     circuit
    -0.66
     IRA
    -0.64
     killer
    -0.64
    POSITIVE LOGITS
     Text
    3.86
    Text
    2.39
     TEXT
    1.94
    text
    1.80
     text
    1.74
     texts
    1.60
    TEXT
    1.52
     Typ
    1.33
     Scroll
    1.19
     textual
    1.16
    Activations Density 0.015%

    No Known Activations