INDEX

Explanations

terms or phrases related to references, such as citations or sources

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Co" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

lihood

-0.74

selves

-0.73

 IMAGES

-0.71

 Wanted

-0.66

glers

-0.66

Reloaded

-0.65

self

-0.65

 STATS

-0.64

 Trou

-0.64

 VIDE

-0.63

POSITIVE LOGITS

verage

1.10

agher

1.09

venant

0.98

asters

0.95

aching

0.93

herent

0.93

ordinate

0.93

erc

0.92

aster

0.92

ord

0.92

Activations Density 0.013%

terms or phrases related to references, such as citations or sources

occurrences of the word "Co" in various contexts

No Comments

No Known Activations