INDEX

Explanations

statements indicating notable observations or insights

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 6-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_6/width_16k/average_l0_133

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.6.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

HasForeignKey

-0.72

 Funny

-0.68

 funny

-0.65

 fhort

-0.64

funny

-0.61

DialogContent

-0.60

Odd

-0.58

Funny

-0.58

 ActivatedRoute

-0.57

 weird

-0.56

POSITIVE LOGITS

μφωνα

0.53

mployed

0.49

 copie

0.49

adaptiveStyles

0.48

balleur

0.47

setzen

0.46

Closer

0.46

 bajos

0.45

jstor

0.45

Activations Density 0.324%

statements indicating notable observations or insights

No Comments

No Known Activations