INDEX

Explanations

expressions of negation or avoidance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

ViewInit

-0.81

InjectAttribute

-0.70

 &___

-0.69

CppMethod

-0.67

 kasarigan

-0.65

 résulte

-0.64

 ProtoMessage

-0.64

quelize

-0.63

ClientSize

-0.63

 Normdatei

-0.63

POSITIVE LOGITS

试试

0.69

 chande

0.67

 kans

0.60

ArgsConstructor

0.59

 Conley

0.59

 Pues

0.58

 favore

0.58

干脆

0.58

atalos

0.58

gettes

0.57

Activations Density 0.053%

expressions of negation or avoidance

No Comments

No Known Activations