INDEX

Explanations

mentions of the word "stress."

oai_token-act-pair · gpt-3.5-turbo

instances of the word "dress" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

©¶æ

-0.73

gum

-0.71

lder

-0.67

 underrated

-0.65

 elig

-0.65

 stockp

-0.64

 defunct

-0.62

 antiv

-0.61

 everlasting

-0.61

vit

-0.61

POSITIVE LOGITS

ively

1.01

ress

1.00

ions

0.99

ional

0.95

ãĥĺ

0.90

ylum

0.90

inator

0.89

mann

0.89

entials

0.87

enger

0.87

Activations Density 0.005%

mentions of the word "stress."

instances of the word "dress" and its variations

No Comments

No Known Activations