INDEX

Explanations

intensifiers or modifiers that amplify the quality of an adjective or adverb

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

Efq

-1.00

 myſelf

-0.96

 Majefty

-0.93

 Houſe

-0.91

 ſche

-0.88

ſelf

-0.88

 vectorstock

-0.86

ſelves

-0.85

 Chriftian

-0.85

 Theſe

-0.85

POSITIVE LOGITS

0.86

 most

0.62

an

0.59

 very

0.58

our

0.56

 extremely

0.55

non

0.55

 another

0.54

0.53

NOSIS

0.53

Activations Density 0.126%

intensifiers or modifiers that amplify the quality of an adjective or adverb

No Comments

No Known Activations