INDEX

Explanations

transitional words indicating the structure of a discourse or argument

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

__["

-0.54

={({

-0.52

 Yosemite

-0.51

ValueGeneration

-0.51

SystemColors

-0.51

ckså

-0.51

 monument

-0.51

 Obsidian

-0.50

Pew

-0.50

WriteAttribute

-0.50

POSITIVE LOGITS

Firstly

1.02

 Firstly

0.95

 firstly

0.88

первых

0.71

 Secondly

0.71

Secondly

0.66

abord

0.63

First

0.62

 Pertama

0.60

 Thirdly

0.60

Activations Density 0.012%

transitional words indicating the structure of a discourse or argument

No Comments

No Known Activations