INDEX

Explanations

words related to negative or critical evaluations and judgments

oai_token-act-pair · gpt-3.5-turbo

adjectives and verbs that describe negative or critical states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ournal

-0.63

 redes

-0.62

 mobilized

-0.62

ITNESS

-0.58

ccording

-0.58

abases

-0.58

 livest

-0.56

 normalized

-0.56

ËĪ

-0.55

 corrid

-0.55

POSITIVE LOGITS

iously

0.85

ious

0.85

ulous

0.80

ively

0.78

uously

0.75

uous

0.73

iveness

0.71

ously

0.71

ingly

0.68

ment

0.68

Activations Density 0.319%

words related to negative or critical evaluations and judgments

adjectives and verbs that describe negative or critical states

No Comments

No Known Activations