INDEX

Explanations

proper nouns related to specific locations or people

oai_token-act-pair · gpt-3.5-turbo

references to specific geographic locations or proper nouns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Magikarp

-0.88

inventory

-0.81

audi

-0.80

alog

-0.78

ramid

-0.76

Brain

-0.73

hib

-0.70

antine

-0.70

cause

-0.70

detail

-0.69

POSITIVE LOGITS

 Wellington

1.02

ienne

0.87

 Oaks

0.86

 Henderson

0.81

ricanes

0.80

 Bulldogs

0.76

iership

0.76

 Island

0.75

 Beach

0.74

igon

0.72

Activations Density 0.005%

proper nouns related to specific locations or people

references to specific geographic locations or proper nouns

No Comments

No Known Activations