INDEX

Explanations

references to people and their statuses or actions in a narrative context

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

/feed

-0.15

 ongoing

-0.14

ceae

-0.14

cis

-0.14

ande

-0.14

 Ð¸Ð¼ÐµÑĤÑĮ

-0.14

Generating

-0.13

attr

-0.13

 Finding

-0.13

iz

-0.13

POSITIVE LOGITS

 going

0.32

going

0.25

 gonna

0.24

 Going

0.23

 coming

0.23

Going

0.22

 joining

0.20

 leaving

0.19

 heading

0.18

 staying

0.17

Activations Density 0.213%

references to people and their statuses or actions in a narrative context

No Comments

No Known Activations