INDEX

Explanations

phrases that indicate recommendations or positive evaluations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_22/width_16k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_mlp_out

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

########.

-0.65

Enllaces

-0.53

SharedCtor

-0.51

#+#

-0.50

setVerticalGroup

-0.49

StoryboardSegue

-0.49

="#"><

-0.47

KeySpec

-0.46

Key

-0.45

 incontra

-0.45

POSITIVE LOGITS

tvguidetime

0.79

GOTREF

0.71

awtextra

0.65

 propOrder

0.64

 ModelExpression

0.59

 typelib

0.57

UALA

0.54

IBOutlet

0.54

ikum

0.53

HasAnnotation

0.51

Activations Density 0.177%

phrases that indicate recommendations or positive evaluations

No Comments

No Known Activations