INDEX

Explanations

concepts related to paradoxes and hypocrisy

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

Äļ

-0.17

uche

-0.15

Ð»Ð°Ð´Ñĥ

-0.14

éĢ¸

-0.14

iÅ¾

-0.14

 Swap

-0.14

ÐŁÐļ

-0.14

 Tape

-0.14

ulong

-0.13

POSITIVE LOGITS

zer

0.15

glas

0.15

sson

0.14

aiser

0.14

 åĨ

0.14

ÙĥØªÙĪØ±

0.14

thern

0.14

.va

0.14

aland

0.14

ober

0.14

Activations Density 0.076%

concepts related to paradoxes and hypocrisy

No Comments

No Known Activations