INDEX

Explanations

adjectives indicating a small or insignificant amount

oai_token-act-pair · gpt-3.5-turbo

the word "mere" in various contexts indicating insignificance or triviality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

lain

-0.78

 Lawyers

-0.75

anwhile

-0.73

 Patriarch

-0.68

 Remastered

-0.66

azel

-0.64

 Nurs

-0.63

 vigorously

-0.63

ACP

-0.62

LCS

-0.62

POSITIVE LOGITS

 mortals

1.10

 guiActiveUn

0.84

reet

0.79

 coincidence

0.76

ifiable

0.76

tons

0.76

ton

0.76

ãĥ¬

0.74

ãĥ©ãĥ³

0.74

 mere

0.72

Activations Density 0.010%

adjectives indicating a small or insignificant amount

the word "mere" in various contexts indicating insignificance or triviality

No Comments

No Known Activations