INDEX

Explanations

quantitative expressions of quantity or scale

oai_token-act-pair · gpt-3.5-turbo

phrases that include fractional quantities or measurements

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

 appropri

-0.66

 alloc

-0.66

 bearer

-0.64

 propri

-0.64

owicz

-0.63

PLA

-0.61

ADA

-0.60

amac

-0.60

 uncond

-0.59

 intrins

-0.59

POSITIVE LOGITS

dozen

0.96

 dozen

0.85

tails

0.84

hops

0.78

 peeled

0.74

ousand

0.72

oil

0.71

agos

0.70

nir

0.69

assed

0.69

Activations Density 0.070%

quantitative expressions of quantity or scale

phrases that include fractional quantities or measurements

No Comments

No Known Activations