INDEX

Explanations

keywords related to the polar regions or polar bears

oai_token-act-pair · gpt-3.5-turbo

references to polar-themed elements or topics

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

AAAA

-0.87

CHA

-0.84

INA

-0.81

LER

-0.79

Rate

-0.78

ITNESS

-0.77

HOME

-0.76

ECH

-0.76

enance

-0.75

eem

-0.74

POSITIVE LOGITS

 polar

1.26

 Polar

0.93

 vortex

0.88

itary

0.87

igr

0.86

 unpopular

0.81

 extremes

0.81

 oppos

0.80

ãĥ£

0.77

 polarized

0.77

Activations Density 0.005%

keywords related to the polar regions or polar bears

references to polar-themed elements or topics

No Comments

No Known Activations