INDEX

Explanations

superlatives and generalizations that emphasize a predominant characteristic

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate general trends or majority opinions

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

heid

-0.90

isa

-0.72

redo

-0.71

acus

-0.69

oak

-0.66

iton

-0.64

atar

-0.60

tis

-0.60

agra

-0.59

arter

-0.58

POSITIVE LOGITS

 mileage

0.69

acci

0.64

 uninterrupted

0.63

igent

0.62

 obvious

0.61

nesota

0.61

 recent

0.61

 comprehensive

0.60

¬¼

0.59

cients

0.58

Activations Density 0.055%

superlatives and generalizations that emphasize a predominant characteristic

phrases that indicate general trends or majority opinions

No Comments

No Known Activations