INDEX

Explanations

occurrences of the term "prime"

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

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

<bos>

-0.66

[...]

-0.65

 Bergh

-0.62

WriteTagHelper

-0.62

GeoNames

-0.60

 McKin

-0.60

 Climat

-0.60

 Castig

-0.60

 Giuli

-0.59

ContextCompat

-0.59

POSITIVE LOGITS

prime

0.64

0.52

SerializeField

0.48

0.43

ResponseBody

0.41

 Gegenstand

0.39

 zijne

0.36

seamnă

0.35

Activations Density 0.022%

occurrences of the term "prime"

No Comments

No Known Activations