INDEX

Explanations

phrases indicating anticipation or expectation

oai_token-act-pair · gpt-3.5-turbo

terms related to anticipation and waiting

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Freak

-0.73

 Quote

-0.61

 inaccur

-0.60

 fluor

-0.60

phys

-0.60

DV

-0.60

 breakup

-0.59

 Stupid

-0.59

Cut

-0.59

 tendency

-0.58

POSITIVE LOGITS

 await

3.64

 awaiting

3.28

 awaits

3.26

 awaited

3.12

awaited

1.52

 waiting

1.50

 waits

1.38

 pending

1.20

 eagerly

1.13

 greeted

1.12

Activations Density 0.020%

phrases indicating anticipation or expectation

terms related to anticipation and waiting

No Comments

No Known Activations