INDEX

Explanations

locations or places described as being near each other

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "near."

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

gear

-0.68

âĶ

-0.66

terior

-0.65

Accessory

-0.64

oleon

-0.63

 arrang

-0.63

ieri

-0.60

BILL

-0.60

["

-0.59

isting

-0.59

POSITIVE LOGITS

by

0.94

sighted

0.87

 Bethlehem

0.87

 Yellowstone

0.82

 Calais

0.82

 Yosemite

0.80

 Damascus

0.79

 where

0.78

 Naples

0.77

 shore

0.76

Activations Density 0.047%

locations or places described as being near each other

occurrences of the word "near."

No Comments

No Known Activations