INDEX

Explanations

quantify the importance or impact of events or concepts

oai_token-act-pair · gpt-3.5-turbo

terms related to significance and importance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ghan

-0.69

ggie

-0.62

abee

-0.61

washer

-0.60

iere

-0.59

 Alone

-0.59

ateurs

-0.59

Zone

-0.58

hire

-0.58

Lay

-0.58

POSITIVE LOGITS

 significance

3.70

 importance

2.19

 relevance

2.12

 implications

1.65

 symbolism

1.54

 signific

1.52

 specificity

1.50

 implication

1.49

 meanings

1.47

 meaning

1.47

Activations Density 0.009%

quantify the importance or impact of events or concepts

terms related to significance and importance

No Comments

No Known Activations