INDEX

Explanations

words related to networks and interconnected systems

oai_token-act-pair · gpt-3.5-turbo

terms related to various types of networks and clusters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 torches

-0.67

 Revelations

-0.63

 Advocate

-0.62

 Fury

-0.60

Tips

-0.59

Nou

-0.57

verbs

-0.57

Tonight

-0.56

ILLE

-0.56

bats

-0.56

POSITIVE LOGITS

 consisting

1.25

 comprised

1.12

 comprising

1.09

 composed

1.01

ipel

0.97

 encomp

0.94

 formed

0.91

of

0.88

 spanning

0.88

 containing

0.86

Activations Density 0.330%

words related to networks and interconnected systems

terms related to various types of networks and clusters

No Comments

No Known Activations