INDEX

Explanations

requests or instructions and emotions related to frustration and anger

oai_token-act-pair · gpt-3.5-turbo

phrases related to confrontation or demanding actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Historically

-0.64

quartered

-0.63

uably

-0.59

 proponents

-0.58

bitious

-0.55

 historically

-0.55

 hallmark

-0.54

itutional

-0.53

ologies

-0.53

 extensively

-0.53

POSITIVE LOGITS

 haha

0.85

-"

0.69

)).

0.68

lol

0.67

âĢ¦."

0.65

!".

0.65

âĢ¦"

0.64

;)

0.64

 fuckin

0.64

..."

0.64

Activations Density 1.297%

requests or instructions and emotions related to frustration and anger

phrases related to confrontation or demanding actions

No Comments

No Known Activations