INDEX

Explanations

references to military locations and institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

IRT

-0.15

åĭ

-0.14

riad

-0.14

 Gins

-0.14

erset

-0.14

 exploration

-0.14

ARI

-0.14

 Exploration

-0.13

ALS

-0.13

stal

-0.13

POSITIVE LOGITS

Compound

0.16

castle

0.15

usters

0.15

Ã¼rk

0.15

 Compound

0.15

 serial

0.14

ofi

0.14

ÃŃÅ¾

0.14

Ú¯Ø§ÙĩÛĮ

0.14

ì¹Ń

0.14

Activations Density 0.034%

references to military locations and institutions

No Comments

No Known Activations