INDEX

Explanations

adverbs or adjectives related to public visibility or disclosure

oai_token-act-pair · gpt-3.5-turbo

instances of the word "publicly."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

nesota

-0.99

nian

-0.77

 Velocity

-0.74

ĸļ

-0.71

vic

-0.70

illary

-0.70

akeru

-0.70

ailand

-0.70

IER

-0.70

 Echo

-0.69

POSITIVE LOGITS

 humiliated

0.87

 traded

0.84

 apologized

0.84

 speaking

0.81

 reprim

0.81

ached

0.79

 acknowledged

0.78

 apologize

0.77

 repud

0.77

 addressed

0.76

Activations Density 0.011%

adverbs or adjectives related to public visibility or disclosure

instances of the word "publicly."

No Comments

No Known Activations