INDEX

Explanations

terms related to negative or harmful consequences and unauthorized actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

 convention

-0.16

izzes

-0.16

eyh

-0.15

 known

-0.15

anian

-0.15

roach

-0.14

counts

-0.14

defs

-0.14

loff

-0.14

scope

-0.14

POSITIVE LOGITS

/problem

0.19

/il

0.19

ities

0.18

ity

0.17

 Practices

0.17

OperationException

0.16

rous

0.16

.Schema

0.16

/to

0.16

 yere

0.16

Activations Density 0.086%

terms related to negative or harmful consequences and unauthorized actions

No Comments

No Known Activations