INDEX

Explanations

specific years referenced in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

mit

-0.17

isko

-0.17

.nlm

-0.16

icon

-0.16

wards

-0.14

igi

-0.14

yna

-0.14

å½±

-0.14

inya

-0.14

æ¶

-0.14

POSITIVE LOGITS

 Aires

0.16

åĭ¢

0.15

icket

0.15

lide

0.13

 ########.

0.13

 Thornton

0.13

iability

0.13

LAS

0.13

strap

0.13

.addObject

0.13

Activations Density 0.023%

specific years referenced in the text

No Comments

No Known Activations