© 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. 20352
    Prev
    Next
    INDEX
    Explanations

    instances where the word "IF" appears

    oai_token-act-pair · gpt-3.5-turbo

    conditional phrases that include "if" and references to specific contexts or conditions

    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
    Russ
    -0.87
    avorite
    -0.78
    upiter
    -0.77
    ophon
    -0.77
    inct
    -0.75
    nick
    -0.75
    enth
    -0.73
    phabet
    -0.72
    asus
    -0.72
    Scot
    -0.72
    POSITIVE LOGITS
     IF
    2.77
     PD
    2.01
     IL
    1.65
     PL
    1.44
     Fas
    1.18
     TL
    1.16
     IS
    1.12
     FF
    1.01
     EL
    1.01
     GF
    0.98
    Activations Density 0.055%

    No Known Activations