INDEX

Explanations

numerical values, particularly those referring to counts or ages of children

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

jong

-0.18

untime

-0.15

ceph

-0.14

ÐµÐ¾

-0.14

 Quote

-0.14

KHR

-0.14

pur

-0.13

ÑĦÑĦ

-0.13

rolled

-0.13

 feeder

-0.13

POSITIVE LOGITS

Interop

0.17

Ø®ÛĮ

0.16

.operator

0.16

ichtig

0.15

.Restr

0.14

eil

0.14

Buk

0.14

ä»¥ä¸Ĭ

0.13

porno

0.13

Activations Density 0.226%

numerical values, particularly those referring to counts or ages of children

No Comments

No Known Activations