INDEX

Explanations

technical terms or jargon related to various scientific fields

oai_token-act-pair · gpt-3.5-turbo

punctuation or symbols associated with data or statistics

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

STL

-0.70

 FIRST

-0.64

 hump

-0.64

izon

-0.62

 Civic

-0.61

 Warehouse

-0.60

 Warp

-0.59

 Chevy

-0.59

 Nether

-0.58

 Cabin

-0.58

POSITIVE LOGITS

bek

0.79

serving

0.69

advertising

0.67

ãĥ´ãĤ¡

0.66

zsche

0.66

Î³

0.65

ket

0.65

Ö¼

0.65

âĶĢ

0.64

theless

0.64

Activations Density 0.047%

technical terms or jargon related to various scientific fields

punctuation or symbols associated with data or statistics

No Comments

No Known Activations