INDEX

Explanations

references to collaboration and teamwork

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 Pride

-0.17

-0.16

ortal

-0.15

edicine

-0.15

ort

-0.14

isions

-0.14

tridge

-0.14

pler

-0.14

POSITIVE LOGITS

 hard

0.20

 according

0.17

hard

0.16

 harder

0.16

-hard

0.15

 upon

0.15

.getHours

0.15

.Fields

0.15

Ø±Ø¨Ùĩ

0.15

 hardest

0.14

Activations Density 0.047%

references to collaboration and teamwork

No Comments

No Known Activations