INDEX

Explanations

phrases that indicate a contrast or exception

oai_token-act-pair · gpt-3.5-turbo

evaluations and opinions about individuals or circumstances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

hiba

-0.65

 sqor

-0.64

cember

-0.62

 Uriel

-0.62

 Ezek

-0.59

roxy

-0.58

 glimps

-0.58

 Quote

-0.57

 backdrop

-0.57

pez

-0.56

POSITIVE LOGITS

 necessarily

1.26

 anymore

1.24

nor

1.20

 anything

1.18

nor

1.11

any

1.10

 EVER

1.05

 slightest

1.00

anything

0.99

ANY

0.96

Activations Density 0.237%

phrases that indicate a contrast or exception

evaluations and opinions about individuals or circumstances

No Comments

No Known Activations