INDEX

Explanations

specific words or types of words

oai_token-act-pair · gpt-3.5-turbo

references to different types of words and their usage in language

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

DERR

-0.95

roxy

-0.78

psey

-0.76

abama

-0.74

cffff

-0.72

ersen

-0.72

©¶æ¥µ

-0.70

etheus

-0.70

taboola

-0.70

rero

-0.69

POSITIVE LOGITS

mith

1.18

sworth

1.06

ptr

0.93

ifier

0.88

words

0.81

 meanings

0.80

 phrases

0.79

 uttered

0.79

processor

0.79

press

0.79

Activations Density 0.052%

specific words or types of words

references to different types of words and their usage in language

No Comments

No Known Activations