INDEX

Explanations

specific company or product names and their mentions in the context of medical research or reports

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-2.48

ekš

-0.73

JKLM

-0.65

vény

-0.65

pośred

-0.62

kozó

-0.58

 שח

-0.58

yszer

-0.57

 الاول

-0.57

 duyg

-0.57

POSITIVE LOGITS

zelfde

0.84

iyle

0.71

ه

0.69

ième

0.68

یتی

0.66

ی

0.64

ený

0.63

sweise

0.59

enchymal

0.59

iqué

0.59

Activations Density 1.158%

specific company or product names and their mentions in the context of medical research or reports

No Comments

No Known Activations