INDEX

Explanations

terms related to journalistic sources

oai_token-act-pair · gpt-3.5-turbo

occurrences of the abbreviation "TO"

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

groups

-0.67

papers

-0.66

busters

-0.66

ifiers

-0.66

minster

-0.66

builders

-0.65

stable

-0.65

mble

-0.64

space

-0.63

 foundations

-0.61

POSITIVE LOGITS

KEN

1.03

OME

0.98

OOL

0.95

OF

0.93

OL

0.92

ilet

0.90

ffee

0.90

YA

0.90

CLASSIFIED

0.89

FU

0.88

Activations Density 0.012%

terms related to journalistic sources

occurrences of the abbreviation "TO"

No Comments

No Known Activations