INDEX

Explanations

adverbs denoting visible or noticeable characteristics

oai_token-act-pair · gpt-3.5-turbo

adverbs indicating visibility or perception

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

enthal

-0.73

spell

-0.72

zsche

-0.71

quote

-0.70

sch

-0.67

bolt

-0.67

berman

-0.65

illet

-0.62

ground

-0.61

orously

-0.60

POSITIVE LOGITS

 noticeable

0.78

 audible

0.78

Decre

0.74

eful

0.73

 twitch

0.70

 absent

0.68

Ju

0.68

Flor

0.67

 Dull

0.65

 visibly

0.65

Activations Density 0.039%

adverbs denoting visible or noticeable characteristics

adverbs indicating visibility or perception

No Comments

No Known Activations