INDEX

Explanations

instances of direct quotations or speech attributed to a higher authority or character

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 denomination

-0.15

Extensions

-0.15

beiten

-0.15

ãĥ¼ãĥĨ

-0.15

 Sens

-0.14

aris

-0.14

.bootstrap

-0.14

andest

-0.14

 Skin

-0.14

arta

-0.13

POSITIVE LOGITS

apest

0.17

0.15

lob

0.15

finger

0.14

ome

0.14

ollo

0.14

fle

0.14

ignum

0.14

UTH

0.14

rimp

0.14

Activations Density 0.084%

instances of direct quotations or speech attributed to a higher authority or character

No Comments

No Known Activations