INDEX

Explanations

expressions related to fractions or percentages

oai_token-act-pair · gpt-3.5-turbo

phrases indicating proportions or fractions of quantities

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

 Cards

-0.71

 Limits

-0.71

ordes

-0.65

ITED

-0.64

 cards

-0.64

Magic

-0.63

Reward

-0.63

Chart

-0.62

Quest

-0.62

 Tycoon

-0.61

POSITIVE LOGITS

ority

0.92

yne

0.77

uous

0.74

emate

0.71

ught

0.69

heric

0.68

omore

0.66

 disapprove

0.66

uously

0.66

pipe

0.66

Activations Density 0.049%

expressions related to fractions or percentages

phrases indicating proportions or fractions of quantities

No Comments

No Known Activations