INDEX

Explanations

mentions of the word "grudge"

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "judge" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

members

-0.78

 listing

-0.67

 Fest

-0.65

 completed

-0.64

Lv

-0.63

 centers

-0.63

 members

-0.63

 ladies

-0.63

 functional

-0.63

val

-0.63

POSITIVE LOGITS

udge

4.52

udging

2.26

ud

1.49

urry

1.23

uzz

1.12

umble

1.10

uge

1.06

uggle

1.01

undle

1.01

iddle

0.97

Activations Density 0.019%

mentions of the word "grudge"

mentions of the word "judge" and its variations

No Comments

No Known Activations