INDEX

Explanations

terms related to medical conditions and anatomical structures

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

ending in "ore" or "ure"

np_acts-logits-general · gemini-2.0-flash

words ending in ure/ore/ere

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 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

 jScrollPane

-0.79

rician

-0.70

isations

-0.69

 AssemblyCulture

-0.68

ohistochemical

-0.67

ApiModelProperty

-0.66

yethylene

-0.66

initComponents

-0.64

enderror

-0.62

🏾

-0.61

POSITIVE LOGITS

Ere

0.49

ns

0.48

ng

0.48

ndra

0.47

 createSlice

0.47

nger

0.46

írito

0.46

search

0.45

 Doreen

0.44

 Throat

0.44

Activations Density 0.159%

terms related to medical conditions and anatomical structures

ending in "ore" or "ure"

words ending in ure/ore/ere

No Comments

No Known Activations