INDEX

Explanations

references to specific locations 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 @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 airports

-0.15

roads

-0.14

ILED

-0.14

ëĮĢë¡ľ

-0.14

 cinemas

-0.14

stores

-0.14

enty

-0.13

-house

-0.13

agan

-0.13

è©

-0.13

POSITIVE LOGITS

 proximity

0.20

 front

0.18

bad

0.18

 prime

0.17

 excellent

0.16

 downtown

0.16

 distance

0.16

 pouring

0.16

 poor

0.16

 walking

0.16

Activations Density 0.193%

references to specific locations or places

No Comments

No Known Activations