INDEX

Explanations

phrases indicating current situations, stakes, and consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ereco

-0.15

ropol

-0.15

.idea

-0.15

-0.14

roman

-0.14

æľĭ

-0.14

 Keto

-0.14

raith

-0.14

ì±Ħ

-0.14

arp

-0.13

POSITIVE LOGITS

FromString

0.14

 sharedApplication

0.14

itzer

0.14

 fals

0.14

untu

0.14

ciler

0.14

nature

0.14

DisplayStyle

0.14

 Else

0.14

æļ®

0.14

Activations Density 0.063%

phrases indicating current situations, stakes, and consequences

No Comments

No Known Activations