INDEX

Explanations

keywords related to datasets

oai_token-act-pair · gpt-3.5-turbo

references to datasets and baseline measurements

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

hammer

-0.80

ition

-0.76

vin

-0.70

ening

-0.70

ertodd

-0.69

ginx

-0.69

eral

-0.69

heart

-0.69

oho

-0.69

fing

-0.69

POSITIVE LOGITS

0.99

ä»£

0.79

Introduced

0.73

mble

0.70

isSpecialOrderable

0.69

icago

0.68

 GOODMAN

0.67

DATA

0.67

 sidx

0.66

Kit

0.66

Activations Density 0.057%

keywords related to datasets

references to datasets and baseline measurements

No Comments

No Known Activations