INDEX

Explanations

phrases indicating the necessity or requirement to do something, such as "need to," "have to," or "had to."

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 intersper

-1.44

 reluct

-1.44

 emphat

-1.43

 thut

-1.41

 depic

-1.40

 shenan

-1.40

 maneu

-1.40

 suspic

-1.39

 vagu

-1.39

 dises

-1.38

POSITIVE LOGITS

be

0.98

 been

0.86

 occur

0.77

<bos>

0.76

 být

0.76

be

0.75

 become

0.74

 come

0.69

 worden

0.68

被

0.67

Activations Density 0.480%

phrases indicating the necessity or requirement to do something, such as "need to," "have to," or "had to."

No Comments

No Known Activations