INDEX

Explanations

time durations that have passed since specific events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "since" indicating elapsed time

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

STAR

-0.70

aves

-0.70

Cu

-0.69

Si

-0.67

ereo

-0.66

å¥³

-0.65

Fal

-0.65

fighter

-0.64

atically

-0.64

bart

-0.63

POSITIVE LOGITS

rely

0.91

 completing

0.80

 encountering

0.75

 arriving

0.73

 elapsed

0.71

 successive

0.70

 conced

0.70

 retiring

0.69

 deciding

0.69

 departed

0.68

Activations Density 0.027%

time durations that have passed since specific events

occurrences of the word "since" indicating elapsed time

No Comments

No Known Activations