INDEX

Explanations

the name "Shi" and its variations across different contexts

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

Tokens starting with "Shi"

np_acts-logits-general · gemini-2.0-flash

Shi followed by mony, math, sh, and others

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

?>">

-0.79

CodedInputStream

-0.77

 /\.(

-0.75

 Exacts

-0.75

 Keating

-0.73

 Linton

-0.73

 Yojana

-0.70

Tikang

-0.68

 oeufs

-0.68

Sach

-0.68

POSITIVE LOGITS

••••

0.72

cible

0.67

PMailer

0.67

XK

0.62

 shim

0.61

Mec

0.61

stri

0.61

ReferenceEquals

0.60

 Bucky

0.60

Рабо

0.60

Activations Density 0.295%

the name "Shi" and its variations across different contexts

Tokens starting with "Shi"

Shi followed by mony, math, sh, and others

No Comments

No Known Activations