INDEX

Explanations

locations or points of origin

oai_token-act-pair · gpt-3.5-turbo

the occurrence of the word "there" in various contexts

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

DAQ

-0.75

 Bund

-0.62

rongh

-0.62

Case

-0.61

HH

-0.60

 recently

-0.60

Ca

-0.58

dylib

-0.58

nr

-0.57

"},"

-0.57

POSITIVE LOGITS

 onwards

1.46

 onward

1.30

forward

1.15

forth

1.01

 sprang

0.87

upon

0.82

 proceeds

0.81

 proceeded

0.80

 proceed

0.78

 arises

0.74

Activations Density 0.042%

locations or points of origin

the occurrence of the word "there" in various contexts

No Comments

No Known Activations