INDEX

Explanations

references to the term "Fat"

oai_token-act-pair · gpt-3.5-turbo

repeated references to the word "Fat."

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

uden

-0.90

incoln

-0.72

cov

-0.71

PLIED

-0.70

espie

-0.69

eele

-0.69

oresc

-0.69

endale

-0.67

roman

-0.66

imbabwe

-0.66

POSITIVE LOGITS

Fat

3.91

Fat

3.08

fat

2.13

FAT

2.04

fat

1.90

 fatty

1.43

 fats

1.27

 Abbas

1.23

 Thin

1.17

Sad

1.14

Activations Density 0.014%

references to the term "Fat"

repeated references to the word "Fat."

No Comments

No Known Activations