INDEX

Explanations

expressions of gratitude and polite conversation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

robe

-0.17

 chant

-0.16

 Gent

-0.16

-0.15

lus

-0.14

ab

-0.14

ing

-0.14

ÑģÐ¸Ð¼

-0.14

agues

-0.14

atta

-0.14

POSITIVE LOGITS

iveau

0.16

-UA

0.16

LOPT

0.15

ìĥ

0.15

ropolis

0.15

.MixedReality

0.14

Ghost

0.14

ivec

0.14

éļı

0.14

ityEngine

0.14

Activations Density 0.041%

expressions of gratitude and polite conversation

No Comments

No Known Activations