INDEX

Explanations

instances of omitted or incomplete information in a context

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

moiselle

-0.73

 Monfieur

-0.73

 Houſe

-0.68

 thyself

-0.66

 myſelf

-0.66

 Diſ

-0.65

 purpoſe

-0.65

 becauſe

-0.64

 pleaſure

-0.63

 poffible

-0.61

POSITIVE LOGITS

expandindo

0.66

Referensi

0.59

AddTagHelper

0.58

<eos>

0.57

 […]

0.57

DropTable

0.57

Lähteet

0.57

 متعلقه

0.56

 ComVisible

0.55

хьтан

0.54

Activations Density 0.079%

instances of omitted or incomplete information in a context

No Comments

No Known Activations