INDEX

Explanations

references to information and its availability or relevance in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uko

-0.20

Ø®Ø§ÙĨÙĩ

-0.17

(es

-0.17

acher

-0.15

Info

-0.15

ongo

-0.15

_Info

-0.15

pty

-0.15

info

-0.15

dogs

-0.15

POSITIVE LOGITS

 about

0.27

ally

0.26

 gathering

0.21

_schema

0.20

 sources

0.20

nal

0.20

/data

0.20

al

0.20

-sharing

0.19

 sheets

0.19

Activations Density 0.073%

references to information and its availability or relevance in various contexts

No Comments

No Known Activations