INDEX

Explanations

healthcare-related terminology and references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ble

-0.14

ulg

-0.14

æ¶²

-0.14

uthor

-0.13

Ara

-0.13

Dismiss

-0.13

acÄ±

-0.13

 ëĺĲ

-0.13

endor

-0.13

ho

-0.13

POSITIVE LOGITS

 celkem

0.16

)((((

0.16

 whistle

0.15

ested

0.15

esi

0.15

etto

0.15

_misc

0.15

ãĥ³ãĥĩ

0.15

RunWith

0.14

_prefix

0.14

Activations Density 0.057%

healthcare-related terminology and references

No Comments

No Known Activations