INDEX

Explanations

mentions of trails

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

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

 trail

-2.27

trail

-1.93

 Trail

-1.84

Trail

-1.72

 trails

-1.48

 TRAIL

-1.35

trails

-1.21

 Trails

-1.17

TRAIL

-1.07

Trails

-1.05

POSITIVE LOGITS

LLocation

0.57

intenant

0.54

bootstrapcdn

0.52

Hentet

0.52

Rumble

0.52

 fusca

0.51

 vagues

0.50

 Taxi

0.50

 Transfermarkt

0.49

 Aguas

0.49

Activations Density 0.176%

mentions of trails

No Comments

No Known Activations