INDEX

Explanations

words related to colors and textures

oai_token-act-pair · gpt-3.5-turbo

mentions of different tones or sound qualities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Assembly

-0.82

ablishment

-0.81

Leaks

-0.81

lund

-0.79

icipated

-0.78

hus

-0.74

drivers

-0.72

UAL

-0.71

idable

-0.71

lisher

-0.71

POSITIVE LOGITS

 Tone

1.16

 tones

0.99

hue

0.97

 tone

0.93

tone

0.88

tones

0.86

 accents

0.84

xual

0.84

 accent

0.79

 tint

0.78

Activations Density 0.023%

words related to colors and textures

mentions of different tones or sound qualities

No Comments

No Known Activations