INDEX

Explanations

numeric values and statistical references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

UnusedPrivate

-0.66

GeneratedCode

-0.64

zenberg

-0.62

Hauptartikel

-0.59

 Italijanski

-0.58

beding

-0.57

 dAtA

-0.56

UseVisualStyle

-0.56

Бахар

-0.55

ⓧ

-0.55

POSITIVE LOGITS

первых

0.65

 Whig

0.60

TagHelper

0.60

etheless

0.56

 Whigs

0.55

queous

0.55

featureID

0.54

 stratigraph

0.54

stander

0.53

 spleen

0.53

Activations Density 0.238%

numeric values and statistical references

No Comments

No Known Activations