INDEX

Explanations

verbs related to understanding or explanation

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate understanding or clarification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ãĤ´ãĥ³

-0.85

pie

-0.75

 benches

-0.73

luster

-0.71

 dips

-0.68

pie

-0.68

bat

-0.67

markets

-0.67

ItemTracker

-0.66

paying

-0.66

POSITIVE LOGITS

hent

0.70

ISE

0.68

 recap

0.65

oca

0.64

arry

0.63

veland

0.62

ealous

0.60

 Shoals

0.59

 Carroll

0.58

 refres

0.58

Activations Density 0.101%

verbs related to understanding or explanation

phrases that indicate understanding or clarification

No Comments

No Known Activations