INDEX

Explanations

numerical expressions indicating a range of values

oai_token-act-pair · gpt-3.5-turbo

phrases relating to margins of error and statistical accuracy

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

stem

-0.74

abies

-0.74

inary

-0.73

 Sparkle

-0.72

dfx

-0.72

bane

-0.72

ruary

-0.71

bris

-0.70

atters

-0.69

prototype

-0.69

POSITIVE LOGITS

cules

0.83

minus

0.78

/-

0.78

++++

0.69

 append

0.67

 infinity

0.67

ï¸

0.66

-+-+-+-+

0.66

 assorted

0.60

 plus

0.60

Activations Density 0.015%

numerical expressions indicating a range of values

phrases relating to margins of error and statistical accuracy

No Comments

No Known Activations