INDEX

Explanations

phrases that indicate product features, benefits, or characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

otr

-0.14

Lac

-0.14

 Triple

-0.14

hdl

-0.14

 rele

-0.13

ouch

-0.13

 ÎľÎ·

-0.13

POSITIVE LOGITS

 Serg

0.15

CD

0.15

rlen

0.15

ijken

0.14

¾

0.14

 Antony

0.14

.proto

0.14

ismet

0.14

ulnerable

0.14

ogo

0.14

Activations Density 0.277%

phrases that indicate product features, benefits, or characteristics

No Comments

No Known Activations