INDEX

Explanations

terms related to a "chain reaction."

oai_token-act-pair · gpt-3.5-turbo

references to "chain reactions" and their implications

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

sburg

-0.89

 "$:/

-0.88

gdala

-0.75

racuse

-0.73

sburgh

-0.72

 Niet

-0.71

vironment

-0.71

odder

-0.69

tk

-0.69

 Gors

-0.69

POSITIVE LOGITS

chain

1.02

 chain

0.92

mail

0.91

Chain

0.87

loader

0.87

wal

0.86

alysis

0.86

worm

0.84

cloth

0.84

breaker

0.81

Activations Density 0.022%

terms related to a "chain reaction."

references to "chain reactions" and their implications

No Comments

No Known Activations