INDEX

Explanations

references to time or events

oai_token-act-pair · gpt-3.5-turbo

recurrent phrases indicating actions or experiences that happen repeatedly

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

 ILCS

-1.01

 Materials

-0.69

pload

-0.69

{:

-0.66

00200000

-0.64

;;;;;;;;;;;;

-0.64

FORMATION

-0.64

asty

-0.63

STON

-0.61

LTD

-0.61

POSITIVE LOGITS

 ever

0.86

 slightest

0.85

 ventured

0.68

 bothered

0.67

 visit

0.66

 anniversary

0.66

inea

0.65

 reincarn

0.65

vana

0.64

cha

0.63

Activations Density 0.238%

references to time or events

recurrent phrases indicating actions or experiences that happen repeatedly

No Comments

No Known Activations