INDEX

Explanations

special characters present in the input text

oai_token-act-pair · gpt-3.5-turbo

references to currency symbols, particularly the pound sterling (£)

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ernels

-0.77

raints

-0.74

ription

-0.73

eanor

-0.70

 Wilmington

-0.70

bleacher

-0.69

ropolitan

-0.69

Esk

-0.68

 suspensions

-0.65

 Sioux

-0.65

POSITIVE LOGITS

£

1.34

®

1.14

\\\\\\\\\\\\\\\\

1.12

å£«

1.04

Ľ

0.95

º

0.94

¾

0.93

çİĭ

0.92

·

0.91

İ

0.90

Activations Density 0.009%

special characters present in the input text

references to currency symbols, particularly the pound sterling (£)

No Comments

No Known Activations