INDEX

Explanations

time-related phrases such as 'within hours', 'within days', 'within minutes'

oai_token-act-pair · gpt-3.5-turbo

references to specific time durations and deadlines

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

imir

-0.77

oided

-0.72

 hypocr

-0.67

acus

-0.62

 Inher

-0.62

 dearly

-0.60

err

-0.60

lav

-0.59

 RAID

-0.59

aped

-0.59

POSITIVE LOGITS

aneously

1.02

 thereafter

0.84

 increments

0.75

 thereof

0.75

ÃĽ

0.71

£ı

0.69

 confines

0.66

 elapsed

0.66

 timeframe

0.65

cms

0.64

Activations Density 0.067%

time-related phrases such as 'within hours', 'within days', 'within minutes'

references to specific time durations and deadlines

No Comments

No Known Activations