INDEX

Explanations

words related to climbing or height

oai_token-act-pair · gpt-3.5-turbo

references to climbing activities or climbers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Ô

-0.82

REDACTED

-0.81

fold

-0.80

 Pacers

-0.74

VERTIS

-0.71

 è£ı

-0.69

 Painter

-0.67

urat

-0.66

Spread

-0.66

DOS

-0.66

POSITIVE LOGITS

 clim

1.14

acter

0.93

actic

0.86

 climbers

0.81

 corrid

0.80

ber

0.77

byss

0.74

atell

0.74

entary

0.73

acus

0.71

Activations Density 0.007%

words related to climbing or height

references to climbing activities or climbers

No Comments

No Known Activations