© 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. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 4098
    Prev
    Next
    INDEX
    Explanations

    Russian text, potentially related to language processing, given the presence of Cyrillic characters and possible vocabulary references

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    462
    +0.12
    0.5%
    326
    +0.11
    0.5%
    103
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    45
    +0.12
    0.03
    31
    +0.11
    0.07
    439
    +0.11
    0.05
    Negative Logits
     GOODMAN
    -0.89
    ometimes
    -0.87
    oppable
    -0.86
    ifact
    -0.84
    emort
    -0.79
    igators
    -0.78
     athlet
    -0.77
    ainer
    -0.75
    itized
    -0.75
    bucks
    -0.73
    POSITIVE LOGITS
    Ñı
    1.34
    н
    1.31
    д
    1.24
    Ñ
    1.20
    ÑĤ
    1.20
    м
    1.17
    л
    1.17
    Ð
    1.16
    ¶
    1.14
    в
    1.12
    Activations Density 4.963%

    No Known Activations