INDEX

Explanations

adverbs followed by adjectives or verbs

oai_token-act-pair · gpt-3.5-turbo

expressions of amazement or emphasis regarding experiences and observations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

odder

-0.67

UME

-0.65

icipated

-0.65

isher

-0.63

 Supplement

-0.63

oubt

-0.62

iculture

-0.62

uthor

-0.61

izu

-0.60

room

-0.59

POSITIVE LOGITS

beit

0.94

 much

0.88

ls

0.88

bill

0.81

lers

0.81

HCR

0.80

much

0.79

 MUCH

0.78

 messed

0.76

pher

0.74

Activations Density 0.077%

adverbs followed by adjectives or verbs

expressions of amazement or emphasis regarding experiences and observations

No Comments

No Known Activations