© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 3-RES-JB
    11. 23610
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences where a sentiment or idea is repeated or endorsed by multiple sources

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "echoed" and related terminology expressing repetition or reflection of sentiments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    amina
    -0.80
    ahime
    -0.75
    aho
    -0.71
    estate
    -0.70
    ole
    -0.70
    enture
    -0.69
    ottest
    -0.69
    talk
    -0.65
    hibition
    -0.64
    erm
    -0.64
    POSITIVE LOGITS
     echoes
    1.12
     echoing
    0.84
     chorus
    0.82
     echoed
    0.78
     reverber
    0.77
     reson
    0.75
     voic
    0.73
     repeats
    0.73
     warnings
    0.72
     repeating
    0.71
    Activations Density 0.014%

    No Known Activations