INDEX

Explanations

words related to technical terms and abbreviations

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to linguistic structures and categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

cember

-0.73

 flats

-0.70

rooms

-0.64

iage

-0.63

çĦ

-0.61

Ô

-0.61

 Samoa

-0.60

 Pigs

-0.59

eele

-0.58

verages

-0.58

POSITIVE LOGITS

andum

0.75

ogyn

0.73

yll

0.70

mal

0.67

alis

0.67

ethy

0.67

GUI

0.65

oidal

0.65

urgical

0.64

thia

0.64

Activations Density 0.113%

words related to technical terms and abbreviations

words and phrases related to linguistic structures and categories

No Comments

No Known Activations