INDEX

Explanations

words that indicate a connection or relation between different entities or situations

oai_token-act-pair · gpt-3.5-turbo

connections and associations between concepts or situations

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

amination

-0.71

ifter

-0.68

estern

-0.67

 Slate

-0.67

odder

-0.67

 Venice

-0.66

Ì¶

-0.65

feeding

-0.64

iami

-0.63

 Selection

-0.62

POSITIVE LOGITS

 relate

1.13

MpServer

0.85

alore

0.83

 challeng

0.82

 relates

0.80

 corrid

0.80

 positively

0.78

ability

0.78

 thereto

0.76

 misunder

0.75

Activations Density 0.008%

words that indicate a connection or relation between different entities or situations

connections and associations between concepts or situations

No Comments

No Known Activations