INDEX

Explanations

target words or phrases

oai_token-act-pair · gpt-3.5-turbo

instances of the word "target."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ansk

-0.71

ooks

-0.69

maid

-0.68

ModLoader

-0.67

loo

-0.65

oline

-0.65

IGH

-0.65

ERROR

-0.63

yg

-0.62

 Expedition

-0.60

POSITIVE LOGITS

ted

1.25

ting

0.83

 range

0.78

ivity

0.77

topic

0.74

izen

0.73

 target

0.70

ched

0.70

rod

0.69

ishes

0.69

Activations Density 0.032%

target words or phrases

instances of the word "target."

No Comments

No Known Activations