INDEX

Explanations

references to documents or records, particularly journals and reports, as well as terms associated with judgment and evaluation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

json

-0.21

joint

-0.20

_json

-0.20

_joint

-0.20

Json

-0.19

/job

-0.18

 Jewish

-0.17

ãĢģæĹ¥æľ¬

-0.17

 judgment

-0.17

 joint

-0.17

POSITIVE LOGITS

venile

0.24

 Báº£n

0.23

ÅĽli

0.22

uxtap

0.21

neys

0.19

stice

0.19

ules

0.19

ifiable

0.19

assic

0.18

uggling

0.18

Activations Density 0.638%

references to documents or records, particularly journals and reports, as well as terms associated with judgment and evaluation

No Comments

No Known Activations