INDEX

Explanations

nouns and phrases that denote positive attributes or endorsements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

 proto

-0.16

Jeh

-0.15

 Proto

-0.15

.gb

-0.14

 Pike

-0.14

 prot

-0.14

blink

-0.14

oucher

-0.14

peat

-0.14

enger

-0.14

POSITIVE LOGITS

assen

0.14

dorf

0.14

 lateral

0.14

abcdefgh

0.13

china

0.13

inar

0.13

Ø§Ø³

0.13

yst

0.13

 heterosexual

0.13

↵↵

0.13

Activations Density 0.075%

nouns and phrases that denote positive attributes or endorsements

No Comments

No Known Activations