INDEX

Explanations

the word "good" with various contexts

oai_token-act-pair · gpt-3.5-turbo

the repeated mention of the word "good."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

âĹ¼

-0.92

Hop

-0.82

eters

-0.76

eds

-0.75

hip

-0.69

agos

-0.68

pper

-0.67

 Pavilion

-0.66

NetMessage

-0.66

udic

-0.66

POSITIVE LOGITS

enough

1.29

 enough

1.00

reads

0.96

sword

0.90

 luck

0.88

 Enough

0.83

 Samar

0.82

 karma

0.81

bye

0.79

 intentions

0.79

Activations Density 0.052%

the word "good" with various contexts

the repeated mention of the word "good."

No Comments

No Known Activations