INDEX

Explanations

phrases that emphasize a specific contrast or comparison context

oai_token-act-pair · gpt-3.5-turbo

referential phrases that lead to explanations or qualifications regarding various subjects

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

uku

-0.77

iol

-0.68

ione

-0.68

any

-0.67

case

-0.65

une

-0.63

yet

-0.63

Family

-0.63

ventions

-0.62

POSITIVE LOGITS

 mattered

1.07

 prompted

0.94

 bothers

0.92

 drew

0.92

 distinguishes

0.90

 inspires

0.89

 attracts

0.87

 drove

0.87

 sparked

0.87

 dominates

0.87

Activations Density 0.116%

phrases that emphasize a specific contrast or comparison context

referential phrases that lead to explanations or qualifications regarding various subjects

No Comments

No Known Activations