INDEX

Explanations

words related to positive attributes or experiences

oai_token-act-pair · gpt-3.5-turbo

expressions of awe or amazement

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

OPLE

-0.89

 Gemini

-0.75

 Luxem

-0.74

cision

-0.70

å£«

-0.69

ãĤ¼

-0.68

cussion

-0.68

 Scha

-0.67

 Schl

-0.67

 Kaepernick

-0.64

POSITIVE LOGITS

akens

1.35

aw

1.04

akening

0.97

atche

0.92

keye

0.91

kward

0.88

aw

0.86

erness

0.86

iring

0.85

ards

0.85

Activations Density 0.004%

words related to positive attributes or experiences

expressions of awe or amazement

No Comments

No Known Activations