INDEX

Explanations

URLs related to specific posts or tweets, likely from social media platforms

oai_token-act-pair · gpt-3.5-turbo

occurrences of URLs or web links

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 compet

-0.65

 shame

-0.64

 exit

-0.63

 advert

-0.61

 arsen

-0.61

 armoured

-0.60

 duck

-0.60

emouth

-0.58

 imperson

-0.57

 pronoun

-0.57

POSITIVE LOGITS

status

1.23

posts

1.07

videos

1.03

comments

1.02

issues

1.01

sets

0.91

photos

0.89

articles

0.87

wiki

0.86

tumblr

0.84

Activations Density 0.026%

URLs related to specific posts or tweets, likely from social media platforms

occurrences of URLs or web links

No Comments

No Known Activations