INDEX

Explanations

words related to news articles and headlines

oai_token-act-pair · gpt-3.5-turbo

occurrences of references to specific dates and events

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Deity

-0.73

 Bowman

-0.67

 Exhibition

-0.64

 Derek

-0.63

eeks

-0.62

 rave

-0.61

 Able

-0.60

 Brandon

-0.60

Cho

-0.60

 Aaron

-0.59

POSITIVE LOGITS

ccording

0.89

Reuters

0.87

Iraq

0.82

Tok

0.82

POL

0.82

WASHINGTON

0.80

externalActionCode

0.78

Prime

0.78

Shares

0.77

AFP

0.76

Activations Density 0.135%

words related to news articles and headlines

occurrences of references to specific dates and events

No Comments

No Known Activations