INDEX

Explanations

terms related to exposure and its potential effects

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

èĨľ

-0.17

browse

-0.16

cing

-0.15

azon

-0.15

reshold

-0.15

lob

-0.14

enge

-0.14

erness

-0.14

lobe

-0.14

.scalablytyped

-0.14

POSITIVE LOGITS

 exposure

0.19

nce

0.17

ively

0.16

 Exposure

0.16

PRINTF

0.16

 ç¤º

0.15

alion

0.15

eyin

0.15

agem

0.15

vation

0.14

Activations Density 0.015%

terms related to exposure and its potential effects

No Comments

No Known Activations