INDEX

Explanations

phrases related to causal relationships, comparisons, and conditions

oai_token-act-pair · gpt-3.5-turbo

words indicating connections, relationships, and influences between concepts or entities

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

ONSORED

-0.72

alion

-0.66

ä»£

-0.62

 Azerb

-0.61

ç«

-0.60

 Vaugh

-0.60

ãĤ¦ãĤ¹

-0.59

 arrang

-0.58

\\\\\\\\

-0.58

 destro

-0.57

POSITIVE LOGITS

uties

0.57

hooting

0.55

 converge

0.51

 Released

0.51

hots

0.49

creen

0.48

criptions

0.48

 unders

0.48

ettings

0.47

 differ

0.47

Activations Density 0.951%

phrases related to causal relationships, comparisons, and conditions

words indicating connections, relationships, and influences between concepts or entities

No Comments

No Known Activations