INDEX

Explanations

verbs followed by the infinitive marker "to."

oai_token-act-pair · gpt-3.5-turbo

instances of the word "to" indicating intention or purpose

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 forbids

-0.62

calling

-0.61

 Bridges

-0.60

 awaiting

-0.59

 newsp

-0.59

Reader

-0.59

blocks

-0.59

 fractions

-0.59

holding

-0.58

 imposing

-0.58

POSITIVE LOGITS

 succeed

1.31

 behave

1.20

be

1.18

 become

1.12

 realise

1.11

 stay

1.07

 flourish

1.06

 come

1.06

 realize

1.06

 disappear

1.02

Activations Density 0.153%

verbs followed by the infinitive marker "to."

instances of the word "to" indicating intention or purpose

No Comments

No Known Activations