INDEX

Explanations

conjunctions and the use of phrases that emphasize continuity or inclusion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 THEN

-0.16

 denen

-0.14

 then

-0.14

IFE

-0.14

 implication

-0.14

msp

-0.14

aso

-0.14

 Then

-0.13

THEN

-0.13

agt

-0.13

POSITIVE LOGITS

is

0.36

has

0.33

can

0.25

was

0.25

 will

0.24

 should

0.23

may

0.21

 ÑıÐ²Ð»ÑıÐµÑĤÑģÑı

0.21

 could

0.20

 although

0.20

Activations Density 0.492%

conjunctions and the use of phrases that emphasize continuity or inclusion

No Comments

No Known Activations