INDEX

Explanations

references to personal care and maintenance products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

lege

-0.15

ÎµÏģÏĮ

-0.15

 Empty

-0.14

StyleSheet

-0.14

EMPTY

-0.14

 redundancy

-0.14

iew

-0.14

aea

-0.14

-0.13

POSITIVE LOGITS

 surfaces

0.38

 delicate

0.21

-sensitive

0.21

 susceptible

0.20

Sensitive

0.20

 sensitive

0.20

 substr

0.20

 unprotected

0.19

 bare

0.19

 porous

0.19

Activations Density 0.197%

references to personal care and maintenance products

No Comments

No Known Activations