INDEX

Explanations

negative phrases and terms related to evaluation or critique

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

sp

-0.63

-0.60

al

-0.59

displaystyle

-0.59

<eos>

-0.58

ba

-0.56

getDefault

-0.56

-0.55

POSITIVE LOGITS

 незавершена

1.34

WithIOException

1.14

 مشين

1.04

 '\\;'

1.02

Билгалдахарш

1.02

]='\

0.99

ніципалі

0.90

 فريبيس

0.89

CopyWith

0.88

 мәкал

0.88

Activations Density 0.003%

negative phrases and terms related to evaluation or critique

No Comments

No Known Activations