INDEX

Explanations

references to similarity or equivalence in contexts discussing amounts, measures, and conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ng

-0.16

ebi

-0.15

Ã¶l

-0.15

anzi

-0.15

ibi

-0.15

à¸£à¸¡

-0.15

inning

-0.15

ka

-0.14

à¥įà¤¸à¤°

-0.14

#Region

-0.14

POSITIVE LOGITS

nock

0.21

ukkan

0.18

0.16

iah

0.14

raig

0.14

 Trav

0.14

JK

0.14

anford

0.14

é¥°

0.14

 nisi

0.13

Activations Density 0.149%

references to similarity or equivalence in contexts discussing amounts, measures, and conditions

No Comments

No Known Activations