INDEX

Explanations

phrases related to problem-solving and instructional queries

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

Code-related actions after "to"

np_acts-logits-general · gemini-2.0-flash

performing an action

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/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.24.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

 chewed

-0.48

 sitting

-0.48

updatedAt

-0.47

avor

-0.46

TypedDataSet

-0.46

onCreate

-0.45

 hành

-0.45

 standing

-0.45

 spends

-0.44

sitting

-0.44

POSITIVE LOGITS

 access

0.76

 proceed

0.64

 retrieve

0.64

InputTagHelper

0.64

 Access

0.60

AccessFile

0.58

 view

0.58

Access

0.57

 activate

0.56

 delete

0.56

Activations Density 0.466%

phrases related to problem-solving and instructional queries

Code-related actions after "to"

performing an action

No Comments

No Known Activations