INDEX

Explanations

contractions and negations

oai_token-act-pair · gpt-3.5-turbo

negative statements and qualifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

arently

-0.72

anwhile

-0.71

 apparently

-0.71

 basically

-0.70

pedia

-0.69

 essentially

-0.68

 presumably

-0.62

ingly

-0.61

 reportedly

-0.61

ALLY

-0.61

POSITIVE LOGITS

gged

0.69

onen

0.67

TAIN

0.65

å¤

0.65

\\\\\\\\

0.65

ciplinary

0.61

ibe

0.57

 enthusi

0.57

olics

0.56

fer

0.56

Activations Density 0.268%

contractions and negations

negative statements and qualifiers

No Comments

No Known Activations