INDEX

Explanations

rankings and comparisons related to size, importance, or position in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

thâu

-0.66

Hochspringen

-0.63

 referrerpolicy

-0.62

abestanden

-0.61

 EconPapers

-0.60

 "..\..\

-0.59

KeySpec

-0.59

 nød

-0.58

-0.57

例句

-0.57

POSITIVE LOGITS

 largest

0.68

largest

0.54

 AssemblyCompany

0.54

 nation

0.52

 fastest

0.51

 Largest

0.50

 ranking

0.48

 longest

0.48

ensively

0.47

Largest

0.47

Activations Density 0.240%

rankings and comparisons related to size, importance, or position in various contexts

No Comments

No Known Activations