INDEX

Explanations

information related to posting content online

oai_token-act-pair · gpt-3.5-turbo

instances of the word "posted."

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

glers

-0.71

kos

-0.68

phant

-0.68

inho

-0.67

 Galile

-0.66

 Gand

-0.66

ichick

-0.64

pter

-0.63

apy

-0.63

FK

-0.63

POSITIVE LOGITS

ulate

0.90

hum

0.89

doctoral

0.88

 postings

0.88

gres

0.84

posted

0.84

uploads

0.83

 posting

0.82

 posts

0.80

 posted

0.75

Activations Density 0.037%

information related to posting content online

instances of the word "posted."

No Comments

No Known Activations