INDEX

Explanations

people or entities mentioned in a situation or event

oai_token-act-pair · gpt-3.5-turbo

references to individuals involved in various situations or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 guiName

-0.74

 Definitions

-0.64

 weakening

-0.63

 Meaning

-0.62

 beginning

-0.62

 melting

-0.62

emphasis

-0.59

 Dictionary

-0.59

steamapps

-0.58

 guiActiveUnfocused

-0.58

POSITIVE LOGITS

 attended

1.25

 participated

1.23

 witnessed

1.17

 visited

1.13

oped

1.10

 accompanied

1.03

 oversaw

1.02

 intervened

1.00

 interacted

1.00

 frequ

0.99

Activations Density 0.128%

people or entities mentioned in a situation or event

references to individuals involved in various situations or actions

No Comments

No Known Activations