INDEX

Explanations

terms related to striking, attacking, or significant actions taken 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 @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

li

-0.49

-0.48

押し

-0.47

 Enti

-0.46

def

-0.45

jint

-0.44

丞

-0.44

 Kurt

-0.42

hp

-0.42

nti

-0.42

POSITIVE LOGITS

 strikes

1.30

 Strikes

1.30

 strike

1.26

Strike

1.22

 Strike

1.21

 Striking

1.18

strike

1.11

 striking

1.09

 struck

1.08

struck

0.97

Activations Density 0.061%

terms related to striking, attacking, or significant actions taken in various contexts

No Comments

No Known Activations