© 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. 22357
    Prev
    Next
    INDEX
    Explanations

    mentions or references to an organization or news agency named "AFP" in the text

    oai_token-act-pair · gpt-3.5-turbo

    references to the news agency AFP

    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
     soph
    -0.67
    Stud
    -0.66
     backward
    -0.63
    ubs
    -0.63
     backwards
    -0.61
     Gems
    -0.61
     hol
    -0.60
    stic
    -0.60
    birds
    -0.59
     crystall
    -0.59
    POSITIVE LOGITS
     AFP
    3.98
    AFP
    2.38
     Reuters
    1.74
     REUTERS
    1.61
     Getty
    1.51
     EPA
    1.36
     AP
    1.21
    Reuters
    1.10
    UTERS
    1.09
     UNHCR
    1.06
    Activations Density 0.010%

    No Known Activations