INDEX

Explanations

references to dogs, specifically their various activities and attributes

oai_token-act-pair · gpt-3.5-turbo

references to dogs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

noon

-0.69

 witness

-0.69

 oral

-0.62

 peninsula

-0.62

 Meier

-0.61

 Falk

-0.60

 Bucc

-0.58

 contact

-0.58

 Rockefeller

-0.58

 sudden

-0.57

POSITIVE LOGITS

ogs

1.27

warts

1.11

glers

1.01

gers

0.90

ogging

0.87

ogg

0.85

zag

0.83

ota

0.80

mire

0.79

asus

0.79

Activations Density 0.004%

references to dogs, specifically their various activities and attributes

references to dogs

No Comments

No Known Activations