INDEX

Explanations

locations or proper nouns

oai_token-act-pair · gpt-3.5-turbo

prominent names and titles in various contexts

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

 challeng

-0.67

 traged

-0.66

_.

-0.65

 disadvant

-0.63

 Vaugh

-0.62

 thereafter

-0.61

jri

-0.61

 thereof

-0.61

atever

-0.61

 disg

-0.60

POSITIVE LOGITS

 âĢº

0.91

 Vegan

0.68

 Updated

0.68

 Calculator

0.65

][

0.61

 Podcast

0.60

Â¶

0.60

 Episode

0.60

 Expand

0.60

 Transcript

0.58

Activations Density 1.316%

locations or proper nouns

prominent names and titles in various contexts

No Comments

No Known Activations