INDEX

Explanations

terms related to prediction and interaction in scientific contexts

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

before methods or outcomes of actions

np_acts-logits-general · gemini-2.0-flash

predict, interact, compare, resist, explain

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 الحره

-0.71

naire

-0.71

iecie

-0.71

lichung

-0.67

PointerException

-0.65

ised

-0.64

ratulations

-0.64

jupiter

-0.64

 Boyer

-0.63

isholm

-0.62

POSITIVE LOGITS

able

1.07

ings

0.86

ingly

0.84

ability

0.73

 pleaſure

0.66

ables

0.65

 InputDecoration

0.64

 themſelves

0.63

pement

0.62

aways

0.61

Activations Density 1.135%

terms related to prediction and interaction in scientific contexts

before methods or outcomes of actions

predict, interact, compare, resist, explain

No Comments

No Known Activations