INDEX

Explanations

references to executive officers and their titles within organizational contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

aption

-0.15

figcaption

-0.15

ect

-0.15

itto

-0.15

strcasecmp

-0.14

 Commercial

-0.14

roi

-0.14

roj

-0.14

bon

-0.14

yme

-0.13

POSITIVE LOGITS

uples

0.17

emand

0.17

Untitled

0.15

opies

0.15

à¸±à¸Ķ

0.14

jÅ¡ÃŃ

0.14

-manager

0.14

æķ¦

0.14

Activations Density 0.006%

references to executive officers and their titles within organizational contexts

No Comments

No Known Activations