INDEX

Explanations

paths and references to user directory structures

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

glDelete

-0.54

kuun

-0.50

pertise

-0.49

śle

-0.48

ENEFITS

-0.48

ioms

-0.48

 cassa

-0.47

 Senado

-0.47

locationManager

-0.47

 represented

-0.47

POSITIVE LOGITS

发表于

0.81

 rospy

0.80

 BrowserModule

0.78

unknownFields

0.77

protoimpl

0.76

CloseOperation

0.74

 Infórmanos

0.72

AndroidJUnit

0.72

 typelib

0.72

 LUMP

0.72

Activations Density 0.031%

paths and references to user directory structures

No Comments

No Known Activations