INDEX

Explanations

choices and decisions made by individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 "..\..\

-0.63

akterysty

-0.61

XtraGrid

-0.60

seteq

-0.56

Fet

-0.56

 Miele

-0.56

 Rutland

-0.55

zsef

-0.55

liminaries

-0.55

 "..\..\..\

-0.55

POSITIVE LOGITS

 décidé

0.85

 opting

0.81

 opted

0.81

 decidió

0.78

 valgt

0.77

 decided

0.77

 decidieron

0.75

حياتها

0.71

GenerationType

0.70

inflater

0.70

Activations Density 0.139%

choices and decisions made by individuals

No Comments

No Known Activations