INDEX

Explanations

words related to beliefs or disbelief, particularly focusing on non-believers

oai_token-act-pair · gpt-3.5-turbo

words related to belief and disbelief

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

displayText

-0.76

ALE

-0.67

Discussion

-0.64

ONS

-0.63

 Extended

-0.61

NES

-0.61

 Pros

-0.60

 eclipse

-0.60

ORPG

-0.60

ãģ¾

-0.60

POSITIVE LOGITS

bel

1.35

ieving

1.22

ievers

1.21

iever

1.15

ieve

1.15

ieves

1.08

ief

1.05

ieved

1.04

iev

0.96

ichick

0.90

Activations Density 0.008%

words related to beliefs or disbelief, particularly focusing on non-believers

words related to belief and disbelief

No Comments

No Known Activations