INDEX

Explanations

occurrences of the word "the"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

åľŃ

-0.17

ï¼ľ

-0.14

nom

-0.14

onen

-0.14

nesty

-0.14

more

-0.13

ãĥĨãĥ«

-0.13

edImage

-0.13

owl

-0.13

orno

-0.13

POSITIVE LOGITS

ses

0.17

OMPI

0.17

ifs

0.17

ÑĪÐµÐ²

0.16

 following

0.16

cko

0.15

pei

0.15

 Following

0.15

 ropes

0.15

oplayer

0.15

Activations Density 0.115%

occurrences of the word "the"

No Comments

No Known Activations