INDEX

Explanations

names or key terms related to people or entities

oai_token-act-pair · gpt-3.5-turbo

proper nouns, specifically names of people

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

 ..."

-0.84

...

-0.73

 âĢ¦"

-0.70

 ....

-0.70

 ......

-0.63

åĤ

-0.61

 guiActiveUnfocused

-0.61

 âĢ¦

-0.60

ãĤ¼ãĤ¦ãĤ¹

-0.57

..."

-0.57

POSITIVE LOGITS

romeda

0.97

anyahu

0.90

espie

0.88

withstanding

0.86

psey

0.85

jamin

0.83

notations

0.83

dinand

0.82

utsche

0.81

odore

0.80

Activations Density 0.221%

names or key terms related to people or entities

proper nouns, specifically names of people

No Comments

No Known Activations