INDEX

Explanations

words related to awe and wonder

oai_token-act-pair · gpt-3.5-turbo

expressions of awe or wonder

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Gemini

-0.79

OPLE

-0.74

 Luxem

-0.73

 Princ

-0.72

å£«

-0.68

 Viet

-0.68

ãĤ¼

-0.65

 Plum

-0.64

ptives

-0.64

 Feld

-0.63

POSITIVE LOGITS

akens

1.32

aw

1.10

akening

1.02

atche

0.98

erness

0.88

kward

0.86

isec

0.84

keye

0.84

awk

0.83

ashington

0.82

Activations Density 0.005%

words related to awe and wonder

expressions of awe or wonder

No Comments

No Known Activations