INDEX

Explanations

instances of the word "when" in statements indicating uncertainty or lack of knowledge

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "when."

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

agin

-0.66

zzi

-0.65

kaya

-0.61

rolet

-0.61

gur

-0.59

bear

-0.59

actor

-0.59

endish

-0.59

elman

-0.59

edly

-0.59

POSITIVE LOGITS

soever

1.31

irlf

0.94

abouts

0.82

 confronted

0.79

 faced

0.73

IPS

0.72

theless

0.70

 pressed

0.66

 asked

0.66

 transitioning

0.65

Activations Density 0.121%

instances of the word "when" in statements indicating uncertainty or lack of knowledge

occurrences of the word "when."

No Comments

No Known Activations