INDEX

Explanations

phrases indicating professional roles or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.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

findpost

-0.70

GRANTED

-0.63

 بيها

-0.61

 WHICH

-0.56

ěstí

-0.55

OGND

-0.55

WriteTagHelper

-0.55

EXIST

-0.55

 Created

-0.54

awtextra

-0.54

POSITIVE LOGITS

 based

0.68

 employed

0.65

 finishing

0.60

 heavily

0.59

ArrowToggle

0.58

encodeWith

0.57

 married

0.56

 involved

0.55

 completing

0.55

 fascinated

0.55

Activations Density 0.201%

phrases indicating professional roles or titles

No Comments

No Known Activations