© 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. 8-RES-JB
    11. 6216
    Prev
    Next
    INDEX
    Explanations

    phrases related to hearing or being informed about something

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "heard about."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     teasp
    -0.84
    alon
    -0.76
    ither
    -0.76
    soever
    -0.75
    earing
    -0.73
    respective
    -0.73
    emo
    -0.72
    arel
    -0.72
    rouse
    -0.71
    ainer
    -0.71
    POSITIVE LOGITS
     Kafka
    0.70
     Vec
    0.67
     them
    0.66
    76561
    0.64
     FB
    0.64
     Opportun
    0.64
    /+
    0.64
     Schwar
    0.63
     waivers
    0.63
     unpublished
    0.62
    Activations Density 0.107%

    No Known Activations