INDEX

Explanations

phrases related to learning and extracting lessons from experiences

oai_token-act-pair · gpt-3.5-turbo

constructions that reference the act of learning from experiences or observations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ahan

-0.66

gui

-0.66

exclusive

-0.65

achable

-0.62

amation

-0.60

 curtains

-0.60

YE

-0.59

owl

-0.58

istent

-0.58

 excluded

-0.58

POSITIVE LOGITS

 firsthand

0.95

Pract

0.84

Format

0.78

 Teach

0.76

DragonMagazine

0.75

ERY

0.74

çīĪ

0.70

format

0.69

 Drill

0.68

 skills

0.67

Activations Density 0.233%

phrases related to learning and extracting lessons from experiences

constructions that reference the act of learning from experiences or observations

No Comments

No Known Activations