INDEX

Explanations

locations and descriptors related to addresses or places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ihn

-0.17

 Ð¸Ð¼Ñĥ

-0.15

riad

-0.15

Elo

-0.15

 villages

-0.14

 gall

-0.14

unks

-0.14

GIN

-0.14

Ð¸Ð·

-0.13

lava

-0.13

POSITIVE LOGITS

 Station

0.20

 Main

0.17

 Church

0.17

 Magazine

0.17

Station

0.16

pliers

0.16

 High

0.16

	extern

0.15

station

0.15

 Narrow

0.15

Activations Density 0.122%

locations and descriptors related to addresses or places

No Comments

No Known Activations