INDEX

Explanations

phrases or conditional statements indicating the ability of subjects to perform actions or achieve outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

Ð°ÑĪ

-0.15

acket

-0.15

uh

-0.15

 Serious

-0.15

aku

-0.14

 Responsible

-0.14

icip

-0.14

orig

-0.14

ahat

-0.14

zi

-0.13

POSITIVE LOGITS

 Ø¨ØªÙĪØ§ÙĨ

0.18

 INTERRU

0.15

can

0.15

 proper

0.15

 accred

0.15

 better

0.14

ITED

0.14

 hopefully

0.14

 ballo

0.13

ouns

0.13

Activations Density 0.095%

phrases or conditional statements indicating the ability of subjects to perform actions or achieve outcomes

No Comments

No Known Activations