INDEX

Explanations

mentiones of the word "toast."

oai_token-act-pair · gpt-3.5-turbo

references to toast as a food item

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

CPC

-0.63

 illegally

-0.63

relation

-0.63

 migration

-0.62

ression

-0.62

rowth

-0.61

 Refugee

-0.59

acqu

-0.59

 Migration

-0.59

bara

-0.59

POSITIVE LOGITS

 toast

3.90

 Toast

2.13

 bread

1.17

 pancakes

1.08

 bacon

0.89

asted

0.88

 fries

0.87

 frying

0.84

 butter

0.83

 parchment

0.80

Activations Density 0.002%

mentiones of the word "toast."

references to toast as a food item

No Comments

No Known Activations