INDEX

Explanations

words related to specific languages and characters from those languages

oai_token-act-pair · gpt-3.5-turbo

symbols or characters that appear in non-standard or special encoding formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 mathemat

-0.84

 treadmill

-0.76

 scrut

-0.75

 incorpor

-0.74

 subsequ

-0.71

 princ

-0.69

 Turing

-0.69

 traged

-0.67

 challeng

-0.67

stadt

-0.66

POSITIVE LOGITS

ï¸ı

1.53

ï¸

1.01

×¨

0.99

ÙĨ

0.99

âĶĢâĶĢ

0.98

×ķ

0.96

ÙĦ

0.96

×ľ

0.95

âķĲâķĲ

0.93

×Ļ×

0.92

Activations Density 0.030%

words related to specific languages and characters from those languages

symbols or characters that appear in non-standard or special encoding formats

No Comments

No Known Activations