INDEX

Explanations

verbs indicating the ability to perform actions or accomplish tasks

oai_token-act-pair · gpt-3.5-turbo

verbs that indicate the ability or action taken by subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

æ©Ł

-0.73

é¾į

-0.70

 wont

-0.67

uesday

-0.64

 shalt

-0.63

nih

-0.63

cember

-0.62

Vert

-0.61

 pitted

-0.61

 followed

-0.60

POSITIVE LOGITS

 safely

0.79

 uninterrupted

0.78

 comfortably

0.78

uate

0.72

successfully

0.70

tein

0.68

 muster

0.66

 satisf

0.66

 convinc

0.65

utenberg

0.65

Activations Density 0.279%

verbs indicating the ability to perform actions or accomplish tasks

verbs that indicate the ability or action taken by subjects

No Comments

No Known Activations