INDEX

Explanations

phrases or sentences where something is being evaluated or judged

oai_token-act-pair · gpt-3.5-turbo

expressions of what is deemed to be true or valuable 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 @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

inas

-0.79

dayName

-0.73

mouth

-0.71

essions

-0.70

ciating

-0.69

hire

-0.68

LET

-0.68

eries

-0.67

ijn

-0.64

ranging

-0.64

POSITIVE LOGITS

phas

0.93

 worthy

0.92

 acceptable

0.89

 taboo

0.89

 synonymous

0.89

 expend

0.87

 harmless

0.87

 unworthy

0.86

 irrelevant

0.85

 advisable

0.85

Activations Density 0.051%

phrases or sentences where something is being evaluated or judged

expressions of what is deemed to be true or valuable in various contexts

No Comments

No Known Activations