INDEX

Explanations

terms related to safety events or conferences

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iginal

-0.15

erras

-0.15

bundle

-0.14

procs

-0.14

 reservoir

-0.14

ÙģØµÙĦ

-0.14

ignal

-0.13

 multiplic

-0.13

nown

-0.13

Subscription

-0.13

POSITIVE LOGITS

 booth

0.48

 exhib

0.45

 Booth

0.42

 booths

0.42

 exhibit

0.41

 exhibits

0.36

 Exhib

0.36

 exhibiting

0.35

 display

0.34

 stall

0.34

Activations Density 0.157%

terms related to safety events or conferences

No Comments

No Known Activations