INDEX

Explanations

specific mentions of the word "thread"

oai_token-act-pair · gpt-3.5-turbo

references to discussion threads

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

kef

-0.67

 plaintiffs

-0.65

 satell

-0.65

emies

-0.64

 Plaint

-0.63

mosp

-0.61

 Flavoring

-0.61

issan

-0.61

urses

-0.61

cÃ©

-0.61

POSITIVE LOGITS

bare

1.51

 thread

1.30

thread

1.27

 threads

1.07

 Thread

1.01

worm

0.91

Thread

0.91

worms

0.91

need

0.89

forum

0.86

Activations Density 0.011%

specific mentions of the word "thread"

references to discussion threads

No Comments

No Known Activations