INDEX

Explanations

instances of adjectives followed by nouns, indicating the evaluation or description of something

oai_token-act-pair · gpt-3.5-turbo

articles and quantifiers indicating a degree of quality or quantity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

anism

-0.74

 Measures

-0.66

 endeavors

-0.66

agree

-0.64

 Lies

-0.60

anners

-0.60

 omin

-0.60

speak

-0.59

 wisely

-0.59

Ec

-0.57

POSITIVE LOGITS

 glimpse

1.18

 foothold

1.14

lot

1.08

 chance

1.00

hold

0.99

 haircut

0.82

bit

0.80

 clue

0.80

 bunch

0.78

 huge

0.77

Activations Density 0.135%

instances of adjectives followed by nouns, indicating the evaluation or description of something

articles and quantifiers indicating a degree of quality or quantity

No Comments

No Known Activations