INDEX

Explanations

places or locations mentioned in a text

oai_token-act-pair · gpt-3.5-turbo

references to various locations or places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

icer

-0.81

TAG

-0.69

olyn

-0.68

quel

-0.67

rd

-0.67

onder

-0.66

ernand

-0.64

CHAT

-0.64

irst

-0.62

vous

-0.62

POSITIVE LOGITS

bos

1.16

holders

1.12

abouts

0.96

holder

0.92

 where

0.88

upon

0.86

 holder

0.78

 else

0.76

 frequ

0.75

beer

0.71

Activations Density 0.074%

places or locations mentioned in a text

references to various locations or places

No Comments

No Known Activations