INDEX

Explanations

phrases indicating a comparison of increasing quantities or intensities

oai_token-act-pair · gpt-3.5-turbo

the repetition of the word "and" indicating a continued list or a buildup of ideas

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

aiden

-0.64

afety

-0.63

digy

-0.63

hoe

-0.58

aturday

-0.58

ONSORED

-0.57

dden

-0.57

edom

-0.55

oola

-0.55

ixie

-0.54

POSITIVE LOGITS

rogens

0.90

 farther

0.84

 more

0.82

 clearer

0.81

 better

0.81

rogen

0.81

 louder

0.78

 stronger

0.75

 faster

0.74

more

0.73

Activations Density 0.027%

phrases indicating a comparison of increasing quantities or intensities

the repetition of the word "and" indicating a continued list or a buildup of ideas

No Comments

No Known Activations