INDEX

Explanations

terms related to evaluation and assessment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 pleaſure

-0.72

 myſelf

-0.70

 Bourgoin

-0.69

 Reſ

-0.67

 purpoſe

-0.67

ſelf

-0.66

 Houſe

-0.65

ſelves

-0.64

UnknownFieldSet

-0.64

 FontWeight

-0.64

POSITIVE LOGITS

دانشنامهٔ

0.71

 resourceCulture

0.47

 quality

0.47

 качество

0.45

mpä

0.45

 good

0.44

 kval

0.43

meer

0.43

iyi

0.42

 petto

0.42

Activations Density 0.719%

terms related to evaluation and assessment

No Comments

No Known Activations