INDEX

Explanations

terms related to confusion and complexity in various contexts

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

confusion

np_acts-logits-general · gemini-2.0-flash

confused, tangled, or messy

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

antMatchers

-0.61

PositiveButton

-0.58

ddot

-0.57

期刊论文

-0.57

 aDecoder

-0.56

ativement

-0.56

lgari

-0.55

mybatisplus

-0.53

ecore

-0.52

UrlResolution

-0.52

POSITIVE LOGITS

 confusing

1.88

 confusion

1.87

 Confusion

1.62

 chaos

1.61

 confused

1.60

 confuse

1.59

Confused

1.56

Confusion

1.56

 confuso

1.53

 confuses

1.49

Activations Density 0.500%

terms related to confusion and complexity in various contexts

confusion

confused, tangled, or messy

No Comments

No Known Activations