INDEX

Explanations

phrases related to connections or associations between different entities or concepts

oai_token-act-pair · gpt-3.5-turbo

references to connections or links between entities or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ĨĴ

-0.84

geist

-0.78

Ĥª

-0.74

©¶æ¥µ

-0.70

ŃĶ

-0.70

 SERV

-0.67

orman

-0.67

iola

-0.67

ihad

-0.66

ctory

-0.66

POSITIVE LOGITS

 shore

0.78

asted

0.75

 conserve

0.75

 facilitate

0.73

 assist

0.72

 accommodate

0.70

Boh

0.70

 ensure

0.68

ethyl

0.67

ggles

0.66

Activations Density 0.144%

phrases related to connections or associations between different entities or concepts

references to connections or links between entities or concepts

No Comments

No Known Activations