INDEX

Explanations

locations or directions related to performing specific actions

oai_token-act-pair · gpt-3.5-turbo

instructional phrases directing users to navigate or access specific resources online

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 flung

-0.67

 hurled

-0.64

 severely

-0.62

 blanket

-0.61

iated

-0.60

 trumpet

-0.60

 nailed

-0.60

 transpl

-0.60

iating

-0.59

 thrown

-0.58

POSITIVE LOGITS

www

1.15

 Settings

1.00

 http

0.95

 https

0.94

 Preferences

0.92

 Menu

0.83

 google

0.82

ebin

0.79

 Pastebin

0.76

 Downloads

0.75

Activations Density 0.140%

locations or directions related to performing specific actions

instructional phrases directing users to navigate or access specific resources online

No Comments

No Known Activations