INDEX

Explanations

occurrences of the word "expected" in relation to output and protocols

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

LookAnd

-0.82

 prét

-0.69

 aveug

-0.68

bkz

-0.67

 ladite

-0.64

 beginnetje

-0.64

 avoient

-0.62

 dizaine

-0.62

Said

-0.59

AddField

-0.59

POSITIVE LOGITS

expected

1.22

 Expected

1.12

Expected

1.11

 expected

1.11

 expectations

1.06

 expectation

1.04

expectations

0.96

expectation

0.95

 expectancy

0.89

EXPECTED

0.87

Activations Density 0.127%

occurrences of the word "expected" in relation to output and protocols

No Comments

No Known Activations