INDEX

Explanations

terms related to long-lasting effects or impressions

oai_token-act-pair · gpt-3.5-turbo

phrases that imply enduring or significant effects

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

ramid

-0.95

aucus

-0.84

idelines

-0.81

eah

-0.80

alon

-0.79

onne

-0.78

umo

-0.76

aucuses

-0.74

olog

-0.73

ologies

-0.73

POSITIVE LOGITS

 lasting

1.08

Effects

0.84

lasting

0.79

 impressions

0.79

 Effects

0.78

 friendships

0.78

 memories

0.75

 pains

0.75

åĬ

0.73

 adolesc

0.73

Activations Density 0.035%

terms related to long-lasting effects or impressions

phrases that imply enduring or significant effects

No Comments

No Known Activations