INDEX

Explanations

negations and phrases indicating exceptions or contrasts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

-0.18

ounder

-0.16

innacle

-0.16

 meanwhile

-0.15

ale

-0.15

Ð¾Ð¿Ð°Ñģ

-0.15

IMA

-0.14

addir

-0.14

imar

-0.14

esco

-0.14

POSITIVE LOGITS

 necessarily

0.17

çĹ

0.16

withstanding

0.16

chers

0.16

ANJI

0.15

ting

0.15

adena

0.14

rons

0.14

 Ð´ÑĢÐµÐ²

0.14

tingham

0.14

Activations Density 0.047%

negations and phrases indicating exceptions or contrasts

No Comments

No Known Activations