INDEX

Explanations

words indicating diversity or variety in offerings

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

emark

-0.15

luetooth

-0.14

ë£Į

-0.14

urious

-0.14

ybrid

-0.14

meer

-0.14

inz

-0.13

 Cotton

-0.13

rr

-0.13

 exact

-0.13

POSITIVE LOGITS

anh

0.16

eyse

0.15

athan

0.14

ÃŃses

0.14

dden

0.14

UNET

0.14

chner

0.14

ETA

0.14

 GenerationType

0.14

isable

0.14

Activations Density 0.012%

words indicating diversity or variety in offerings

No Comments

No Known Activations