INDEX

Explanations

references to game shows and competitive activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

DY

-0.15

.synthetic

-0.14

ukkit

-0.14

illet

-0.14

 inserts

-0.14

ë³µì§Ģ

-0.14

dration

-0.14

DRV

-0.14

 Stops

-0.14

iaux

-0.14

POSITIVE LOGITS

aggio

0.15

èī¯

0.15

organic

0.14

Ð¾ÑĢÐ°Ð·

0.14

èĪį

0.14

ners

0.14

amu

0.14

oser

0.14

.module

0.14

wt

0.14

Activations Density 0.289%

references to game shows and competitive activities

No Comments

No Known Activations