INDEX

Explanations

instances of future events or predictions regarding actions or outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 olab

-0.17

ç«Ł

-0.16

 Ð²Ð¾Ð¾Ð±ÑīÐµ

-0.15

hin

-0.15

 Sometimes

-0.15

ÑģÑĸÐ¼

-0.14

âĦĸâĦĸ

-0.14

íĭ±

-0.14

.Packet

-0.14

Ã²n

-0.14

POSITIVE LOGITS

 likely

0.94

 probably

0.82

likely

0.81

 Likely

0.76

probably

0.72

 probable

0.68

 Probably

0.68

Probably

0.65

 most

0.63

 likelihood

0.62

Activations Density 0.682%

instances of future events or predictions regarding actions or outcomes

No Comments

No Known Activations