INDEX

Explanations

specifically highlighted phrases

oai_token-act-pair · gpt-3.5-turbo

words and phrases that emphasize specificity and detail

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

00200000

-0.66

 hurd

-0.65

 dexter

-0.63

arthed

-0.62

usercontent

-0.61

hou

-0.59

ricks

-0.59

cox

-0.57

":[{"

-0.56

 satur

-0.56

POSITIVE LOGITS

anship

0.87

 Levant

0.67

tainment

0.63

hedon

0.62

 insofar

0.62

 Kindle

0.61

ports

0.60

_>

0.59

âĢ¦)

0.58

 MacArthur

0.56

Activations Density 0.255%

specifically highlighted phrases

words and phrases that emphasize specificity and detail

No Comments

No Known Activations