INDEX

Explanations

words containing the suffix "-ble"

oai_token-act-pair · gpt-3.5-turbo

words that convey a sense of ambiguity or vagueness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Yuk

-0.63

 includ

-0.63

ans

-0.61

 calendars

-0.60

Ans

-0.60

 prosecut

-0.59

 Uran

-0.58

 linem

-0.58

 advising

-0.57

 forces

-0.57

POSITIVE LOGITS

ble

4.56

bles

2.76

bled

2.61

bling

2.54

BLE

2.47

bly

1.93

blers

1.90

bler

1.65

Ble

1.61

bley

1.60

Activations Density 0.006%

words containing the suffix "-ble"

words that convey a sense of ambiguity or vagueness

No Comments

No Known Activations