INDEX

Explanations

disclaimers or legal notices

oai_token-act-pair · gpt-3.5-turbo

statements related to disclaimers, copyright, and non-affiliation notices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 crunch

-0.77

 piled

-0.75

 catapult

-0.73

miah

-0.71

 estab

-0.69

 dracon

-0.68

 enriched

-0.68

 leap

-0.67

 creeping

-0.67

 morp

-0.67

POSITIVE LOGITS

 Please

1.61

 Therefore

1.39

 PLEASE

1.33

 Accordingly

1.28

Please

1.27

Any

1.26

 However

1.26

 Unless

1.23

If

1.20

 Additionally

1.17

Activations Density 0.258%

disclaimers or legal notices

statements related to disclaimers, copyright, and non-affiliation notices

No Comments

No Known Activations