INDEX

Explanations

words related to posts or postings online

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "post."

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

ãĥīãĥ©

-0.87

xon

-0.73

 Quinn

-0.71

ptive

-0.70

 Berk

-0.67

kson

-0.65

pload

-0.65

ggies

-0.65

veyard

-0.62

ught

-0.62

POSITIVE LOGITS

algia

1.27

een

1.10

rophe

1.03

rand

1.03

ensibly

1.01

ream

1.01

alg

1.01

rophic

0.97

eering

0.97

ricting

0.95

Activations Density 0.020%

words related to posts or postings online

occurrences of the term "post."

No Comments

No Known Activations