INDEX

Explanations

the word "dumped" in various contexts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "dumped" and its variations, indicating themes of rejection or disposal

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

HCR

-0.85

mental

-0.78

autical

-0.75

osuke

-0.72

umer

-0.72

 Flavoring

-0.71

jc

-0.68

iasm

-0.67

tein

-0.67

iola

-0.65

POSITIVE LOGITS

dump

1.22

 dumps

1.19

 dumped

1.13

 dumping

1.03

 dump

0.99

 heap

0.88

liga

0.79

 piles

0.78

loads

0.75

 corrid

0.71

Activations Density 0.006%

the word "dumped" in various contexts

instances of the word "dumped" and its variations, indicating themes of rejection or disposal

No Comments

No Known Activations