INDEX

Explanations

phrases related to the early days or beginnings of various activities or events

oai_token-act-pair · gpt-3.5-turbo

references to temporal phrases, particularly related to "early days."

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

ais

-0.68

itsu

-0.66

aths

-0.65

iu

-0.65

ifi

-0.64

ographically

-0.62

Baltimore

-0.61

compl

-0.61

ãģĹ

-0.61

psc

-0.60

POSITIVE LOGITS

pring

0.90

hift

0.80

 waning

0.79

imester

0.73

poons

0.70

hower

0.69

heet

0.67

cape

0.66

opian

0.66

nings

0.65

Activations Density 0.054%

phrases related to the early days or beginnings of various activities or events

references to temporal phrases, particularly related to "early days."

No Comments

No Known Activations