INDEX

Explanations

words related to continuing education or professional training

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

fin

-0.17

fone

-0.17

 prim

-0.16

/world

-0.16

Cup

-0.15

 VÃ¡c

-0.14

arak

-0.14

anchise

-0.14

bane

-0.14

cup

-0.14

POSITIVE LOGITS

 PATCH

0.15

 Heller

0.15

Ñľ

0.15

quil

0.14

bout

0.14

 Å¡t

0.14

Éµ

0.13

izer

0.13

 Patch

0.13

_wp

0.13

Activations Density 0.012%

words related to continuing education or professional training

No Comments

No Known Activations