INDEX

Explanations

phrases indicating possibility, frequency, or conditionality 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 @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

basicConfig

-0.58

гән

-0.55

 ovviamente

-0.54

essentially

-0.53

DebuggerStep

-0.52

Климат

-0.51

 essentially

-0.47

rootScope

-0.47

 aina

-0.46

 yaptığı

-0.45

POSITIVE LOGITS

 might

1.12

may

1.06

might

1.04

anskje

1.03

 Vielleicht

0.99

 misschien

0.97

 vielleicht

0.96

 maybe

0.95

Might

0.95

 possibly

0.93

Activations Density 1.030%

phrases indicating possibility, frequency, or conditionality in various contexts

No Comments

No Known Activations