INDEX

Explanations

phrases related to legal matters and government officials

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

 Muhamma

-0.60

vek

-0.55

 BIBSYS

-0.55

 moza

-0.55

umo

-0.54

Embeddable

-0.53

 karton

-0.53

 kask

-0.52

 kule

-0.51

cupertino

-0.51

POSITIVE LOGITS

<bos>

1.12

Anything

0.63

 jemals

0.60

Autoritní

0.60

anything

0.58

 whatsoever

0.57

 ANYTHING

0.57

 slightest

0.56

soever

0.56

iconque

0.55

Activations Density 0.303%

phrases related to legal matters and government officials

No Comments

No Known Activations