INDEX

Explanations

instances of locations or places

oai_token-act-pair · gpt-3.5-turbo

the preposition "in" indicating various contexts or locations

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

SPONSORED

-0.87

 antioxid

-0.86

RESULTS

-0.77

 Seym

-0.74

âĢ¢âĢ¢âĢ¢âĢ¢

-0.74

Balt

-0.72

ROR

-0.71

 millenn

-0.71

 cryst

-0.70

VIDEO

-0.70

POSITIVE LOGITS

in

1.26

on

0.76

 within

0.76

 during

0.75

 inside

0.74

in

0.74

at

0.73

 throughout

0.70

 therein

0.69

 under

0.68

Activations Density 0.694%

instances of locations or places

the preposition "in" indicating various contexts or locations

No Comments

No Known Activations