INDEX

Explanations

references to year-end summaries and recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

assen

-0.17

idel

-0.16

olie

-0.16

 genders

-0.15

uta

-0.15

 instant

-0.15

avic

-0.15

olan

-0.14

 Hide

-0.14

oka

-0.14

POSITIVE LOGITS

pio

0.16

æĸ

0.15

thin

0.14

cion

0.14

ning

0.14

nothrow

0.14

Intermediate

0.14

 intermediate

0.14

mans

0.14

lue

0.13

Activations Density 0.046%

references to year-end summaries and recommendations

No Comments

No Known Activations