INDEX

Explanations

citations and bibliographic references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Table

-0.15

andi

-0.15

 Right

-0.15

panse

-0.14

 bach

-0.14

 Entry

-0.14

 List

-0.14

 Directors

-0.14

/Table

-0.14

usercontent

-0.14

POSITIVE LOGITS

doi

0.28

note

0.24

 journal

0.23

doi

0.23

.publisher

0.22

 publisher

0.22

 note

0.22

keywords

0.21

publisher

0.21

journal

0.21

Activations Density 0.005%

citations and bibliographic references

No Comments

No Known Activations