INDEX

Explanations

negations and terms that express a lack of validity or legitimacy

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

lero

-0.16

ALER

-0.15

arts

-0.14

xes

-0.14

Ŀ

-0.14

Ð½Ð°ÑĤ

-0.14

asons

-0.14

'gc

-0.14

elyn

-0.14

iyon

-0.14

POSITIVE LOGITS

 Trot

0.17

Ø¨ÙĪØ§Ø³Ø·Ø©

0.15

Ø±Ø§Ùĩ

0.14

coni

0.14

uala

0.14

ìŀ¥ìĿĦ

0.14

 Crossing

0.14

ermann

0.13

osl

0.13

Activations Density 0.005%

negations and terms that express a lack of validity or legitimacy

No Comments

No Known Activations