INDEX

Explanations

logical conditions and checks in programming structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

τσι

-0.47

 deepcopy

-0.46

 quarters

-0.44

 abri

-0.43

 uncut

-0.42

Naissance

-0.41

reht

-0.40

ugia

-0.40

VolleyError

-0.39

malink

-0.38

POSITIVE LOGITS

Is

1.52

Is

1.35

Has

1.24

is

1.15

Has

1.14

IsContent

1.07

IsIn

0.99

IsInitialized

0.97

is

0.97

IS

0.95

Activations Density 0.282%

logical conditions and checks in programming structures

No Comments

No Known Activations