INDEX

Explanations

expressions related to avoidance or evasion of negative situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

fter

-0.15

/Dk

-0.15

enal

-0.15

ileo

-0.15

 Levin

-0.14

hm

-0.14

 Nisan

-0.14

à¤®à¤°

-0.14

atura

-0.14

enha

-0.13

POSITIVE LOGITS

ance

0.20

/mit

0.16

aleigh

0.16

odge

0.15

ably

0.14

enger

0.14

tex

0.14

rette

0.14

reno

0.14

/loose

0.14

Activations Density 0.024%

expressions related to avoidance or evasion of negative situations

No Comments

No Known Activations