INDEX

Explanations

Japanese characters

oai_token-act-pair · gpt-3.5-turbo

symbols or special characters typically seen in non-standard text or encoding contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ossier

-0.77

 eleph

-0.71

 Dragonbound

-0.71

 rupture

-0.68

 Franch

-0.67

ovych

-0.66

 similarities

-0.66

oulos

-0.65

 delegates

-0.65

ocamp

-0.63

POSITIVE LOGITS

¤

1.74

ĺ

1.68

¹

1.67

Ķ

1.65

¥

1.62

İ

1.61

´

1.59

Ĩ

1.58

Ħ

1.58

Ĳ

1.56

Activations Density 0.031%

Japanese characters

symbols or special characters typically seen in non-standard text or encoding contexts

No Comments

No Known Activations