INDEX

Explanations

business-related terms and phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

-0.17

raya

-0.16

 Lauderdale

-0.16

 involved

-0.14

lava

-0.14

riel

-0.14

nock

-0.13

 ÄĮech

-0.13

 Sparks

-0.13

 ÐºÐ°Ð´

-0.13

POSITIVE LOGITS

nist

0.17

ade

0.16

au

0.16

iras

0.14

OLID

0.14

_userdata

0.14

Ngh

0.14

yscale

0.14

 pathMatch

0.14

ioni

0.14

Activations Density 1.533%

business-related terms and phrases

No Comments

No Known Activations