INDEX

Explanations

time-related expressions related to particular durations or occurrences like days, weeks, weekends, nights, and years

oai_token-act-pair · gpt-3.5-turbo

references to time periods, specifically days, weeks, months, and years

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

emort

-0.83

 channels

-0.74

responsible

-0.71

wrong

-0.68

 heirs

-0.68

ioch

-0.67

 guiIcon

-0.66

bidden

-0.65

haar

-0.63

ortium

-0.62

POSITIVE LOGITS

dream

0.86

mar

0.77

break

0.76

days

0.73

night

0.73

iversary

0.73

adan

0.71

ivan

0.70

 opener

0.69

mary

0.68

Activations Density 0.144%

time-related expressions related to particular durations or occurrences like days, weeks, weekends, nights, and years

references to time periods, specifically days, weeks, months, and years

No Comments

No Known Activations