INDEX

Explanations

numerical factors or ratings

oai_token-act-pair · gpt-3.5-turbo

terms related to contributing factors or influences 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

quartered

-0.79

escription

-0.79

gem

-0.77

 Seym

-0.77

alach

-0.77

kay

-0.76

©¶æ

-0.75

gements

-0.74

phis

-0.74

ccording

-0.74

POSITIVE LOGITS

 factor

0.92

factor

0.86

 factors

0.83

inity

0.80

 equation

0.76

imate

0.72

witz

0.71

 Factors

0.71

 Factor

0.69

iator

0.68

Activations Density 0.017%

numerical factors or ratings

terms related to contributing factors or influences in various contexts

No Comments

No Known Activations