INDEX

Explanations

timestamps or references to current events in articles

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the present continuous state or ongoing events

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

onnaissance

-0.88

 Pant

-0.67

phasis

-0.66

oute

-0.66

andom

-0.64

anwhile

-0.64

 Dragonbound

-0.62

 Asset

-0.61

 Anat

-0.61

 Ange

-0.60

POSITIVE LOGITS

 housed

0.98

 belonged

0.92

 referred

0.91

 represented

0.88

 attained

0.88

 encountered

0.87

©¶æ

0.86

 contained

0.86

 constituted

0.86

 held

0.86

Activations Density 0.223%

timestamps or references to current events in articles

phrases indicating the present continuous state or ongoing events

No Comments

No Known Activations