INDEX

Explanations

references to chronological events or historical information

oai_token-act-pair · gpt-3.5-turbo

elements of significant changes or events in historical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Yesterday

-0.81

terday

-0.71

ĪĴ

-0.71

resa

-0.65

 Accessed

-0.65

umbn

-0.64

Ability

-0.64

abus

-0.64

 QUEST

-0.64

POS

-0.64

POSITIVE LOGITS

 thereafter

1.37

 afterward

1.15

 subsequent

1.15

 culminating

1.13

 eventually

1.11

 afterwards

1.05

Eventually

1.03

 Eventually

1.02

 until

1.00

 later

0.98

Activations Density 0.645%

references to chronological events or historical information

elements of significant changes or events in historical contexts

No Comments

No Known Activations