INDEX

Explanations

phrases indicating simplicity or ease of action

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

-0.77

-0.67

-0.65

<i>

-0.60

<em>

-0.60

-0.59

ra

-0.59

-0.58

fun

-0.57

POSITIVE LOGITS

easily

1.38

Easily

1.26

)";

1.25

 Easily

1.23

 fácilmente

1.18

 easily

1.17

 unknownFields

1.16

 typelib

1.15

 مرئيه

1.13

'},

1.11

Activations Density 0.070%

phrases indicating simplicity or ease of action

No Comments

No Known Activations