INDEX

Explanations

phrases indicating assumptions and hypothetical scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 venons

-0.74

 Bracken

-0.70

MultipartFile

-0.65

 attiv

-0.65

ynchro

-0.63

Ba

-0.63

Tri

-0.61

zha

-0.60

fcntl

-0.60

 Werken

-0.60

POSITIVE LOGITS

 assume

1.72

 assumes

1.56

assume

1.51

 Assumptions

1.49

 assuming

1.45

 assumed

1.42

 assum

1.41

 Assume

1.40

Assume

1.37

 assumptions

1.36

Activations Density 0.187%

phrases indicating assumptions and hypothetical scenarios

No Comments

No Known Activations