INDEX

Explanations

references to the word "Zero" followed by a number which can go up to 10

oai_token-act-pair · gpt-3.5-turbo

references to the term "Zero" 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

owship

-0.87

amental

-0.84

ammad

-0.82

sbm

-0.80

ogical

-0.79

ular

-0.78

lished

-0.78

igious

-0.74

ument

-0.73

atural

-0.73

POSITIVE LOGITS

MQ

0.97

Zero

0.86

 Zero

0.81

 Threat

0.76

 Gravity

0.76

 Hedge

0.75

 Tone

0.74

Tier

0.73

 Escape

0.72

 Dwell

0.71

Activations Density 0.045%

references to the word "Zero" followed by a number which can go up to 10

references to the term "Zero" in various contexts

No Comments

No Known Activations