INDEX

Explanations

phrases that convey collaborative relationships and connections between subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

vet

-0.18

evi

-0.15

agra

-0.15

 Grim

-0.15

ving

-0.15

ç´ł

-0.15

orthand

-0.14

 Norris

-0.14

wet

-0.14

alsy

-0.14

POSITIVE LOGITS

idlo

0.17

 çĻº

0.16

leet

0.14

dio

0.14

jÃŃm

0.14

.habbo

0.14

UDIO

0.14

 voks

0.14

olor

0.14

Activations Density 0.385%

phrases that convey collaborative relationships and connections between subjects

No Comments

No Known Activations