INDEX

Explanations

keywords related to mud and dirty environments

oai_token-act-pair · gpt-3.5-turbo

references to mud or muddy conditions

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

ECT

-0.73

zona

-0.73

ASED

-0.72

Sax

-0.72

ATURE

-0.68

å£«

-0.67

OPLE

-0.66

trl

-0.64

Recommend

-0.62

ENCE

-0.62

POSITIVE LOGITS

geon

0.92

bath

0.87

bag

0.86

alore

0.85

urai

0.83

iquette

0.80

hog

0.78

sl

0.77

du

0.75

ucer

0.75

Activations Density 0.011%

keywords related to mud and dirty environments

references to mud or muddy conditions

No Comments

No Known Activations