INDEX

Explanations

negative statements or denials

oai_token-act-pair · gpt-3.5-turbo

expressions that deny or negate statements

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

anwhile

-0.69

 Brief

-0.63

ALLY

-0.63

 progressively

-0.62

ationally

-0.62

 essentially

-0.62

 basically

-0.61

psey

-0.61

ĻĤ

-0.60

ingly

-0.60

POSITIVE LOGITS

onen

0.70

å¤

0.67

 underest

0.63

ibe

0.63

spir

0.63

von

0.61

ettings

0.60

omething

0.60

 enthusi

0.59

gged

0.59

Activations Density 0.194%

negative statements or denials

expressions that deny or negate statements

No Comments

No Known Activations