INDEX

Explanations

instances of sarcasm or irony in language

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ces

-0.16

itle

-0.16

apan

-0.15

mdat

-0.15

ppv

-0.15

itz

-0.14

ellan

-0.14

adesh

-0.14

unar

-0.14

oston

-0.14

POSITIVE LOGITS

dns

0.17

wis

0.16

dns

0.15

.rl

0.14

Î·ÏĤ

0.14

trail

0.14

esti

0.14

Č↵

0.14

 dear

0.14

 Trails

0.14

Activations Density 0.698%

instances of sarcasm or irony in language

No Comments

No Known Activations