INDEX

Explanations

references to collective concepts or actions

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 itſelf

-1.08

 pleaſure

-1.01

ſelf

-0.96

 reaſon

-0.95

脚注の使い方

-0.94

 Theſe

-0.93

 <<<<<<<<<<<<<<

-0.92

 NSCoder

-0.92

 auffi

-0.91

ſelves

-0.90

POSITIVE LOGITS

Collective

0.62

 directed

0.61

 defeated

0.56

 collective

0.55

 contest

0.54

 incomplete

0.54

 Poliz

0.53

 prioritize

0.53

Selective

0.53

 defeat

0.52

Activations Density 0.089%

references to collective concepts or actions

No Comments

No Known Activations