INDEX

Explanations

words that indicate causality or implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

SBATCH

-0.66

رشف

-0.66

 Cæsar

-0.66

 sight

-0.66

 Bela

-0.65

 agrarian

-0.64

 Gher

-0.63

 utafitiHapana

-0.63

 cactus

-0.61

 Shakspeare

-0.61

POSITIVE LOGITS

所以在

0.65

 demek

0.60

 many

0.59

So

0.59

HasForeignKey

0.58

oporosis

0.58

 بنابراین

0.58

といって

0.58

so

0.57

HideInInspector

0.57

Activations Density 0.318%

words that indicate causality or implications

No Comments

No Known Activations