INDEX

Explanations

names of individuals, specifically those relevant to the context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

 Plate

-0.17

 sunk

-0.15

criptors

-0.14

upo

-0.14

 kans

-0.14

vÄĽÅĻ

-0.14

icity

-0.14

erge

-0.14

/Instruction

-0.14

 raise

-0.14

POSITIVE LOGITS

::-

0.16

ackson

0.15

 Andrew

0.15

 stake

0.15

0.14

.AppCompatActivity

0.14

Ø´Ø©

0.13

irth

0.13

Andrew

0.13

Activations Density 0.031%

names of individuals, specifically those relevant to the context

No Comments

No Known Activations