INDEX

Explanations

references to programming environments and error messages

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

 <=",

-0.79

 مشين

-0.72

insee

-0.66

Халык

-0.64

 HasFactory

-0.63

 iſt

-0.61

 Numerade

-0.60

 فريبيس

-0.60

 AssemblyCulture

-0.58

:✨

-0.58

POSITIVE LOGITS

1.18

‑

0.81

‐

0.77

־

0.74

−

0.63

0.61

$-

0.60

0.55



0.54

–

0.51

Activations Density 10.851%

references to programming environments and error messages

No Comments

No Known Activations