INDEX

Explanations

expressions related to assumptions or conclusions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

DockStyle

-0.75

RegressionTest

-0.74

 queſta

-0.70

HomeAsUpEnabled

-0.67

ftagPool

-0.63

TagMode

-0.61

StructEnd

-0.60

StreetMap

-0.60

UnknownFieldSet

-0.59

 stockbild

-0.57

POSITIVE LOGITS

be

2.25

 быть

1.06

 been

1.01

 være

1.00

 być

0.99

 essere

0.94

 être

0.93

 have

0.91

 belong

0.90

 být

0.90

Activations Density 1.258%

expressions related to assumptions or conclusions

No Comments

No Known Activations