INDEX

Explanations

elements related to locations and their connections

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

ÏĤ

-0.16

emet

-0.16

 INTO

-0.16

 into

-0.15

 Relative

-0.15

Relative

-0.14

uma

-0.14

 note

-0.14

emente

-0.14

Ta

-0.14

POSITIVE LOGITS

0.17

Ã´le

0.16

ptions

0.16

hazi

0.15

ilver

0.15

à¸±à¸Ļà¸Ķ

0.15

ope

0.14

tÄĽ

0.14

nings

0.14

éĤ£éĩĮ

0.14

Activations Density 0.070%

elements related to locations and their connections

No Comments

No Known Activations