INDEX

Explanations

phrases denoting a large quantity or number of something

oai_token-act-pair · gpt-3.5-turbo

quantifiers or terms indicating frequency and quantity

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

istan

-0.69

ESCO

-0.67

Constructed

-0.66

nature

-0.65

axis

-0.60

cause

-0.58

idden

-0.58

KE

-0.58

idd

-0.57

SR

-0.56

POSITIVE LOGITS

 incrim

0.77

 valuable

0.76

 additional

0.75

 amounts

0.72

 meaningful

0.71

 useful

0.70

 interesting

0.69

 worthwhile

0.69

 impressive

0.69

 amount

0.68

Activations Density 0.341%

phrases denoting a large quantity or number of something

quantifiers or terms indicating frequency and quantity

No Comments

No Known Activations