INDEX

Explanations

words related to locations or places

oai_token-act-pair · gpt-3.5-turbo

references to the word "junior" and related connections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Panic

-0.79

ngth

-0.75

Winner

-0.71

beit

-0.68

senal

-0.68

isable

-0.67

 stocked

-0.66

istor

-0.65

ogun

-0.65

liquid

-0.65

POSITIVE LOGITS

jun

1.54

cture

1.22

ctions

1.00

iors

1.00

 junction

0.94

eteenth

0.87

iper

0.82

eties

0.80

jun

0.79

ames

0.76

Activations Density 0.006%

words related to locations or places

references to the word "junior" and related connections

No Comments

No Known Activations