INDEX

Explanations

instances of the word "to" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

celebrate

-0.68

Celebrate

-0.67

valuate

-0.64

UniformLocation

-0.59

Evaluate

-0.55

δό

-0.53

sume

-0.53

Answered

-0.53

impose

-0.52

atify

-0.52

POSITIVE LOGITS

 hopefully

0.74

 basically

0.70

 actually

0.69

 CreateTagHelper

0.69

 lenker

0.68

 essentially

0.67

not

0.67

 really

0.65

RenderAtEndOf

0.65

 suddenly

0.64

Activations Density 0.411%

instances of the word "to" in various contexts

No Comments

No Known Activations