INDEX

Explanations

verbs related to understanding

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing the concept of understanding

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

onies

-0.83

nar

-0.69

 notor

-0.66

hire

-0.60

pmwiki

-0.60

 roller

-0.60

 bounced

-0.60

 injunction

-0.60

Torrent

-0.60

 sleeves

-0.59

POSITIVE LOGITS

WHY

0.97

 Understanding

0.91

how

0.91

why

0.90

ledge

0.84

 Understand

0.83

ably

0.82

why

0.77

 understanding

0.77

ances

0.76

Activations Density 0.038%

verbs related to understanding

phrases emphasizing the concept of understanding

No Comments

No Known Activations