INDEX

Explanations

terms related to ties or tied situations

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

󠁢

-0.66

 ioutil

-0.66

processable

-0.63

rawDesc

-0.62

 kasarigan

-0.62

amarin

-0.62

FieldNumber

-0.62

tagext

-0.61

 الحره

-0.60

ucket

-0.59

POSITIVE LOGITS

 tied

1.09

tie

1.04

 tying

0.95

 Ties

0.93

 Tied

0.91

 ties

0.89

tied

0.81

Tie

0.77

TIE

0.75

Tie

0.74

Activations Density 0.067%

terms related to ties or tied situations

No Comments

No Known Activations