INDEX

Explanations

references to organizations, applications, and educational settings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

 another

-0.16

iverse

-0.16

 somewhere

-0.16

aso

-0.15

 often

-0.15

aset

-0.15

 indeed

-0.15

-0.14

 punct

-0.14

 perhaps

-0.14

POSITIVE LOGITS

 ONLY

0.19

except

0.17

 except

0.17

_except

0.17

pls

0.16

æĿ¥è¯´

0.16

 domic

0.16

stru

0.15

gnore

0.15

 Only

0.15

Activations Density 0.192%

references to organizations, applications, and educational settings

No Comments

No Known Activations