INDEX

Explanations

locations and actions involving movement or returning

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

 perhatian

-0.51

χρι

-0.51

voke

-0.47

åt

-0.46

orite

-0.46

assertThrows

-0.46

 permanence

-0.45

 dành

-0.45

 crackdown

-0.45

tente

-0.45

POSITIVE LOGITS

 feeling

1.12

 knowing

1.11

 armed

0.92

 wearing

0.92

feeling

0.89

 refreshed

0.89

 clutching

0.88

knowing

0.84

 smelling

0.84

Knowing

0.82

Activations Density 0.354%

locations and actions involving movement or returning

No Comments

No Known Activations