INDEX

Explanations

action verbs indicating the use of something for a specific purpose

oai_token-act-pair · gpt-3.5-turbo

instances of the word "to" across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Known

-0.69

 Caller

-0.66

 Needs

-0.65

ukemia

-0.64

listed

-0.63

 Bhar

-0.63

nil

-0.61

ippi

-0.60

soDeliveryDate

-0.60

tions

-0.59

POSITIVE LOGITS

 promote

1.29

 justify

1.27

 maximize

1.21

 enhance

1.21

 create

1.17

 lure

1.17

 propel

1.16

 bolster

1.14

 generate

1.14

 facilitate

1.14

Activations Density 0.154%

action verbs indicating the use of something for a specific purpose

instances of the word "to" across various contexts

No Comments

No Known Activations