INDEX

Explanations

statements indicating expert opinions and beliefs

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

obox

-0.15

lef

-0.14

unk

-0.14

ropoda

-0.14

mot

-0.14

arket

-0.14

rink

-0.13

ester

-0.13

ivariate

-0.13

éĽĦ

-0.13

POSITIVE LOGITS

 themselves

0.25

 alike

0.16

ìĩ

0.16

orca

0.15

 Olympics

0.15

ordova

0.14

 Kidd

0.14

uma

0.14

 thems

0.14

Activations Density 0.151%

statements indicating expert opinions and beliefs

No Comments

No Known Activations