INDEX

Explanations

prepositions used in relation to specific actions or purposes

oai_token-act-pair · gpt-3.5-turbo

instances of the word "to" and variations with high usage in context

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

natureconservancy

-0.69

 endorsements

-0.65

browser

-0.64

 tracts

-0.62

 interf

-0.58

Gap

-0.58

 hairc

-0.57

 poised

-0.57

ðŁĺ

-0.57

 senses

-0.55

POSITIVE LOGITS

othy

0.85

LEASE

0.84

pload

0.84

 commemorate

0.80

 clarify

0.78

oult

0.77

 preserve

0.75

pper

0.75

sole

0.73

 compensate

0.73

Activations Density 0.308%

prepositions used in relation to specific actions or purposes

instances of the word "to" and variations with high usage in context

No Comments

No Known Activations