INDEX

Explanations

markers of ongoing dialogue or coding constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 Kelley

-0.50

Seg

-0.48

 string

-0.47

 réserve

-0.46

ſer

-0.46

emb

-0.46

ſy

-0.45

 ober

-0.45

resh

-0.45

uyên

-0.44

POSITIVE LOGITS

ंदीखरीदारी

0.81

SequentialGroup

0.81

 jScrollPane

0.76

 pinulongan

0.74

 صوتيه

0.73

 '\\;'

0.72

#+#

0.70

mybatisplus

0.70

 متعلقه

0.70

tagHelperRunner

0.70

Activations Density 0.031%

markers of ongoing dialogue or coding constructs

No Comments

No Known Activations