INDEX

Explanations

references to specific locations or institutions

oai_token-act-pair · gpt-3.5-turbo

the repetition of the word "more."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Legions

-0.70

DD

-0.64

²¾

-0.62

etimes

-0.62

 brightest

-0.61

ARS

-0.60

riber

-0.60

ANGE

-0.60

 corners

-0.60

ĳ

-0.58

POSITIVE LOGITS

theless

0.96

etheless

0.93

than

0.88

pedia

0.86

avid

0.83

more

0.80

agascar

0.75

cam

0.74

likely

0.74

withstanding

0.72

Activations Density 0.013%

references to specific locations or institutions

the repetition of the word "more."

No Comments

No Known Activations