INDEX

Explanations

words related to tents

oai_token-act-pair · gpt-3.5-turbo

references to tents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Words

-0.67

bye

-0.67

MIT

-0.65

obbies

-0.64

 Cuomo

-0.62

History

-0.62

focused

-0.62

FW

-0.61

tracks

-0.60

Export

-0.60

POSITIVE LOGITS

 tent

1.30

acle

1.15

atively

1.13

pole

1.12

 encamp

1.09

 tents

1.08

 flap

0.93

 pole

0.91

 poles

0.87

acles

0.86

Activations Density 0.010%

words related to tents

references to tents

No Comments

No Known Activations