INDEX

Explanations

sequences of numbers, particularly those that include serious medical or data-related references

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

 للاسماء

-1.19

 resourceCulture

-1.12

 '\\;'

-1.11

setVerticalGroup

-1.09

+#+#

-1.03

 itſelf

-1.02

Efq

-0.99

ItemBackground

-0.98

 للمعارف

-0.98

 ModelExpression

-0.97

POSITIVE LOGITS

int

0.62

ra

0.61

led

0.61

El

0.60

AND

0.57

</i>

0.56

0.55

bar

0.54

 others

0.54

Activations Density 0.796%

sequences of numbers, particularly those that include serious medical or data-related references

No Comments

No Known Activations