INDEX

Explanations

references and citations in text

oai_token-act-pair · gpt-3.5-turbo

references or citations within the document

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

 heights

-0.72

daq

-0.71

################

-0.68

uliffe

-0.66

 whiff

-0.66

 deed

-0.64

hawk

-0.64

CHO

-0.62

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

-0.62

adena

-0.62

POSITIVE LOGITS

eree

1.33

erences

1.20

lection

1.19

inement

1.15

lections

1.14

actor

1.14

erred

1.14

erential

1.14

riger

1.12

eren

1.11

Activations Density 0.009%

references and citations in text

references or citations within the document

No Comments

No Known Activations