INDEX

Explanations

statistics or numbers that have exceeded or surpassed a certain threshold

oai_token-act-pair · gpt-3.5-turbo

terms related to surpassing or exceeding thresholds or measures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

atto

-0.78

ale

-0.70

rain

-0.69

bh

-0.68

iott

-0.67

spot

-0.65

BUG

-0.63

abet

-0.62

hya

-0.61

onna

-0.61

POSITIVE LOGITS

 expectations

1.23

 expectation

0.74

 predictions

0.74

 precon

0.66

 usual

0.66

stood

0.65

 anything

0.65

ceptions

0.65

İĭ

0.64

 theirs

0.64

Activations Density 0.109%

statistics or numbers that have exceeded or surpassed a certain threshold

terms related to surpassing or exceeding thresholds or measures

No Comments

No Known Activations