INDEX

Explanations

information about various news stories or political events, possibly with a focus on controversy or conflict

oai_token-act-pair · gpt-3.5-turbo

instances of a specific character or symbol in text

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

 avoidance

-0.67

 reflex

-0.64

matic

-0.64

 lapse

-0.64

 apes

-0.62

 pigeon

-0.61

 avenues

-0.59

ulative

-0.59

pid

-0.59

 Antar

-0.59

POSITIVE LOGITS

ï¸ı

1.01

ever

0.87

arthed

0.84

âĶĢâĶĢâĶĢâĶĢ

0.83

âĶĢâĶĢ

0.81

conom

0.80

ield

0.79

âĤ¬

0.78

uthor

0.77

reci

0.74

Activations Density 0.187%

information about various news stories or political events, possibly with a focus on controversy or conflict

instances of a specific character or symbol in text

No Comments

No Known Activations