INDEX

Explanations

phrases related to comparisons or judgments of quality

oai_token-act-pair · gpt-3.5-turbo

phrases describing various types of entities or characteristics associated with an "a" and "the."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oiler

-0.60

Interstitial

-0.54

double

-0.52

%%

-0.50

 Impossible

-0.50

below

-0.48

slow

-0.48

ļéĨĴ

-0.48

 lesser

-0.47

 Missing

-0.47

POSITIVE LOGITS

 adherents

0.65

 lawy

0.62

 curric

0.58

iership

0.55

 arrivals

0.54

itect

0.52

().

0.52

 endeav

0.51

 existence

0.51

igion

0.51

Activations Density 1.600%

phrases related to comparisons or judgments of quality

phrases describing various types of entities or characteristics associated with an "a" and "the."

No Comments

No Known Activations