INDEX

Explanations

occurrences of the word "on."

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

 InputDecoration

-0.82

Rüyada

-0.81

 virginity

-0.68

ThroughAttribute

-0.64

RectangleBorder

-0.63

 UIFont

-0.62

 pleaſure

-0.61

 tachy

-0.61

 myſelf

-0.61

 isoto

-0.60

POSITIVE LOGITS

 based

0.90

Based

0.79

 Based

0.78

 berdasarkan

0.77

based

0.72

 basada

0.72

@",

0.68

基于

0.68

 BASED

0.66

 основе

0.64

Activations Density 0.179%

occurrences of the word "on."

No Comments

No Known Activations