INDEX

Explanations

references to organizations or entities identified by acronyms or initials

oai_token-act-pair · gpt-3.5-turbo

occurrences of the abbreviations "OR" and "IN" within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

flies

-0.81

uffs

-0.79

unia

-0.77

geons

-0.74

ãĤª

-0.72

ffen

-0.72

rief

-0.71

avia

-0.70

imore

-0.69

angel

-0.69

POSITIVE LOGITS

ER

1.25

TERN

1.21

NESS

1.21

PUT

1.20

CEPT

1.18

IAN

1.16

LY

1.11

IAS

1.11

IST

1.10

ROR

1.09

Activations Density 0.093%

references to organizations or entities identified by acronyms or initials

occurrences of the abbreviations "OR" and "IN" within the text

No Comments

No Known Activations