INDEX

Explanations

terms related to rediscovery or restoration of things

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

istrovstvÃŃ

-0.17

stown

-0.17

ingham

-0.16

.Restr

-0.15

edor

-0.15

edar

-0.14

inea

-0.14

sterdam

-0.14

ÃĶNG

-0.14

zbek

-0.14

POSITIVE LOGITS

uyu

0.16

åŀ

0.16

Chi

0.15

Chi

0.15

ÙĨØ´

0.14

chi

0.14

 integerValue

0.14

indle

0.14

_VOL

0.14

à¸łà¸²à¸©

0.14

Activations Density 0.001%

terms related to rediscovery or restoration of things

No Comments

No Known Activations