INDEX

Explanations

symbols and repeated sequences that indicate formatting or special punctuation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 Ð½Ð¾Ð¶

-0.14

fra

-0.14

lias

-0.14

 ÙĨØ³

-0.14

ersed

-0.14

itori

-0.14

vents

-0.14

#undef

-0.13

legate

-0.13

overlay

-0.13

POSITIVE LOGITS

 holding

0.25

 bearing

0.23

 bear

0.23

 Bear

0.22

bear

0.21

sal

0.21

 Holding

0.21

Bear

0.20

 bears

0.20

 carrying

0.20

Activations Density 0.010%

symbols and repeated sequences that indicate formatting or special punctuation

No Comments

No Known Activations