INDEX

Explanations

phrases indicating a point of view or argument

oai_token-act-pair · gpt-3.5-turbo

verbs indicating actions or events related to commentary

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Higher

-0.77

Versions

-0.74

selves

-0.73

aroo

-0.72

mint

-0.71

OTAL

-0.71

 millenn

-0.71

illion

-0.69

luster

-0.68

illions

-0.67

POSITIVE LOGITS

 herself

0.92

 adamant

0.77

 himself

0.77

 Tsarnaev

0.71

coy

0.70

 arra

0.67

 pandemonium

0.65

 defiant

0.61

 confid

0.59

 Bust

0.58

Activations Density 0.469%

phrases indicating a point of view or argument

verbs indicating actions or events related to commentary

No Comments

No Known Activations