INDEX

Explanations

keywords related to revealing new information or findings

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the act of uncovering or disclosing information

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

etheless

-0.86

jab

-0.78

ortunately

-0.73

 captcha

-0.71

phrine

-0.70

otor

-0.70

yip

-0.69

agues

-0.69

 subur

-0.68

rel

-0.66

POSITIVE LOGITS

 reveals

1.03

 secrets

0.85

 contradictions

0.84

 showc

0.80

orial

0.80

ource

0.80

 reveal

0.80

iary

0.79

ibility

0.76

advertisement

0.75

Activations Density 0.011%

keywords related to revealing new information or findings

phrases indicating the act of uncovering or disclosing information

No Comments

No Known Activations