INDEX

Explanations

phrases indicating a beginning or debut

oai_token-act-pair · gpt-3.5-turbo

the word "top" in various contexts, particularly emphasizing its significance

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

 tender

-0.77

bard

-0.72

schild

-0.70

thur

-0.69

afe

-0.68

enhagen

-0.68

 disarm

-0.64

 affordability

-0.64

ÃĥÃĤ

-0.63

iously

-0.63

POSITIVE LOGITS

rique

0.82

////////////////////////////////

0.81

anol

0.72

rition

0.72

cellence

0.66

ufact

0.66

vg

0.63

lihood

0.62

 nause

0.61

arning

0.60

Activations Density 0.000%

phrases indicating a beginning or debut

the word "top" in various contexts, particularly emphasizing its significance

No Comments

No Known Activations