INDEX

Explanations

the word "only" to indicate exclusivity or uniqueness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

andin

-0.15

 instead

-0.15

Rey

-0.15

antt

-0.14

rab

-0.14

utable

-0.14

rey

-0.14

iland

-0.13

bling

-0.13

IM

-0.13

POSITIVE LOGITS

zos

0.16

tam

0.14

isper

0.14

ellan

0.14

niej

0.14

jez

0.14

amax

0.14

apus

0.14

uzzer

0.13

 svaz

0.13

Activations Density 0.014%

the word "only" to indicate exclusivity or uniqueness

No Comments

No Known Activations