INDEX

Explanations

phrases related to news events and media reports

oai_token-act-pair · gpt-3.5-turbo

mentions of media reports or coverage related to significant events or controversies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

":"/

-0.87

utical

-0.86

Anyway

-0.85

 Anyway

-0.76

 everlasting

-0.73

endif

-0.70

eous

-0.69

 endif

-0.67

natureconservancy

-0.63

ciation

-0.62

POSITIVE LOGITS

 allegations

1.04

 leaked

1.02

 alleged

0.97

 alleging

0.96

 allegedly

0.95

 mistakenly

0.94

 inaccur

0.91

 incorrectly

0.88

 suspected

0.88

 accusing

0.87

Activations Density 0.459%

phrases related to news events and media reports

mentions of media reports or coverage related to significant events or controversies

No Comments

No Known Activations