INDEX

Explanations

proper nouns related to places or names

oai_token-act-pair · gpt-3.5-turbo

references to "Rocky" and "Rocky Mountain"

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

*/(

-0.95

lessly

-0.86

istics

-0.79

usc

-0.79

ogl

-0.78

ugu

-0.72

istically

-0.72

inen

-0.71

IST

-0.71

ihar

-0.70

POSITIVE LOGITS

 Mountain

0.87

 Mountains

0.84

cliffe

0.80

 Helmet

0.79

 Dreams

0.78

 River

0.78

 Horror

0.76

 Romero

0.76

 Mount

0.75

bottom

0.75

Activations Density 0.016%

proper nouns related to places or names

references to "Rocky" and "Rocky Mountain"

No Comments

No Known Activations