© 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. 0-RES-JB
    11. 16826
    Prev
    Next
    INDEX
    Explanations

    words related to technology and digital communications

    oai_token-act-pair · gpt-3.5-turbo

    references to interface elements or technical terms related to programming or data structures

    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
     Zucker
    -0.62
     Grimm
    -0.61
     borough
    -0.60
     categor
    -0.60
     Welsh
    -0.59
     bere
    -0.59
     amph
    -0.58
     ranked
    -0.58
     slide
    -0.57
     NHS
    -0.57
    POSITIVE LOGITS
    INT
    4.44
     INT
    1.89
    INTER
    1.79
    int
    1.65
    EXT
    1.52
    INC
    1.34
    Int
    1.31
    IN
    1.30
    inters
    1.28
    ULT
    1.26
    Activations Density 0.016%

    No Known Activations