INDEX

Explanations

intensifiers and adverbs that indicate degree or comparison

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.17.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

 barrister

-0.54

 deceleration

-0.51

ordning

-0.48

 grossa

-0.47

earlier

-0.47

wahlen

-0.46

 geçen

-0.46

 proposée

-0.45

 vasca

-0.44

 fiches

-0.44

POSITIVE LOGITS

ValueStyle

0.94

NUMX

0.79

tvguidetime

0.77

 المعيارى

0.69

extAlignment

0.69

ContentAsync

0.68

queryInterface

0.65

 تضيفلها

0.64

lillah

0.63

存于互联网档案馆

0.62

Activations Density 0.237%

intensifiers and adverbs that indicate degree or comparison

No Comments

No Known Activations