INDEX

Explanations

terms related to reports, recommendations, and decision-making

oai_token-act-pair · gpt-3.5-turbo

references to reports and recommendations in a formal context

oai_token-act-pair · gpt-4o-mini Triggered 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

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hid

-0.70

grab

-0.68

tein

-0.68

 brilliantly

-0.68

 cult

-0.66

 wonderfully

-0.65

 literal

-0.63

gling

-0.63

 wandered

-0.63

Born

-0.63

POSITIVE LOGITS

 recommendations

1.07

 deliberations

1.06

 findings

0.95

 feasibility

0.92

 finalized

0.91

 consultation

0.89

 conclusions

0.87

 review

0.87

 drafting

0.86

 final

0.85

Activations Density 0.682%

terms related to reports, recommendations, and decision-making

references to reports and recommendations in a formal context

No Comments

No Known Activations