INDEX

Explanations

phrases related to quantitative measures or amounts

oai_token-act-pair · gpt-3.5-turbo

references to counts or amounts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

stal

-0.80

inav

-0.80

hurst

-0.73

gotten

-0.72

robe

-0.71

vironments

-0.66

oneliness

-0.65

doms

-0.65

bones

-0.64

 Doodle

-0.61

POSITIVE LOGITS

 quantity

0.96

 quantities

0.96

Quantity

0.84

quant

0.84

ickr

0.77

 quotas

0.72

 thereof

0.71

tery

0.68

 glut

0.67

ester

0.67

Activations Density 0.015%

phrases related to quantitative measures or amounts

references to counts or amounts

No Comments

No Known Activations