INDEX

Explanations

phrases related to a specific concept or object being mentioned, often in a formal or detailed context

oai_token-act-pair · gpt-3.5-turbo

the word "this."

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

Balt

-0.62

pmwiki

-0.61

APE

-0.57

VIDEO

-0.57

ONS

-0.56

MORE

-0.56

ELS

-0.54

Interstitial

-0.54

AGE

-0.53

PATH

-0.53

POSITIVE LOGITS

 this

2.27

this

1.78

 THIS

1.65

 these

1.65

these

1.29

 THESE

1.14

 tonight

1.13

 This

1.11

This

1.10

THIS

1.03

Activations Density 0.223%

phrases related to a specific concept or object being mentioned, often in a formal or detailed context

the word "this."

No Comments

No Known Activations