INDEX

Explanations

phrases expressing emphasis or contradiction

oai_token-act-pair · gpt-3.5-turbo

negations or expressions of denial

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 eleph

-1.04

 pione

-1.03

ò

-1.01

aditional

-0.98

ortunately

-0.97

 metic

-0.96

Þ

-0.94

ö

-0.92

 practition

-0.92

ą

-0.91

POSITIVE LOGITS

't

1.70

Â´

1.03

\'

0.93

uts

0.89

ÃŃ

0.86

0.85

ï¿½

0.84

Õ

0.77

Ì¶

0.73

bryce

0.72

Activations Density 0.114%

phrases expressing emphasis or contradiction

negations or expressions of denial

No Comments

No Known Activations