INDEX

Explanations

phrases indicating a lack of knowledge or understanding

oai_token-act-pair · gpt-3.5-turbo

expressions of uncertainty or lack of knowledge

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

semble

-0.77

nette

-0.74

visor

-0.72

inka

-0.71

cedes

-0.71

rup

-0.71

ĪĴ

-0.70

ammers

-0.70

conservancy

-0.70

raint

-0.68

POSITIVE LOGITS

 whatsoever

1.07

èª

0.81

 squat

0.78

 guessing

0.71

 guesses

0.70

how

0.66

 whats

0.66

UTH

0.65

 WATCHED

0.64

 glimps

0.64

Activations Density 0.026%

phrases indicating a lack of knowledge or understanding

expressions of uncertainty or lack of knowledge

No Comments

No Known Activations