INDEX

Explanations

phrases indicative of limitations or exceptions in a context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

éľĬ

-0.16

lech

-0.15

ileÅŁ

-0.15

ÏĦÎ¹Ïĥ

-0.15

/styles

-0.15

AXB

-0.15

lico

-0.14

amacare

-0.14

ruk

-0.14

]={↵

-0.14

POSITIVE LOGITS

 benefits

0.19

 initially

0.19

 benefit

0.19

 convenience

0.19

 Benefit

0.18

 relief

0.18

 Benefits

0.18

 choice

0.18

 benef

0.18

 benefited

0.17

Activations Density 0.016%

phrases indicative of limitations or exceptions in a context

No Comments

No Known Activations