INDEX

Explanations

terms related to medical research and study methodology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Whilst

-0.68

Whilst

-0.65

our

-0.59

 whilst

-0.56

 navidad

-0.53

 noël

-0.52

非常的

-0.52

 Ministre

-0.49

 veramente

-0.48

 possu

-0.47

POSITIVE LOGITS

 snippetHide

0.77

 queſta

0.71

ſelf

0.70

 ſche

0.62

 Anſ

0.59

 watchdog

0.59

ſelves

0.59

 lawmakers

0.59

 amid

0.59

 ―――――

0.58

Activations Density 1.098%

terms related to medical research and study methodology

No Comments

No Known Activations