© 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. 1-RES-JB
    11. 16317
    Prev
    Next
    INDEX
    Explanations

    technical terms or codes related to technology or classification labels

    oai_token-act-pair · gpt-3.5-turbo

    references to "cod" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ultras
    -0.69
     Lafayette
    -0.67
     Prosper
    -0.67
     boycott
    -0.66
     pity
    -0.66
     Zimbabwe
    -0.66
    olicy
    -0.65
     Oaks
    -0.65
    oples
    -0.64
    nder
    -0.63
    POSITIVE LOGITS
     cod
    1.26
    Cod
    1.14
    ependent
    1.06
    cod
    1.04
    ãĥĥãĥī
    0.99
     Cod
    0.92
    ename
    0.92
    epend
    0.87
    breaker
    0.86
    ependence
    0.84
    Activations Density 0.003%

    No Known Activations