INDEX

Explanations

comparative phrases, particularly using the word "as" followed by an adjective or adverb

oai_token-act-pair · gpt-3.5-turbo

phrases indicating comparison or similarity

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

ople

-0.67

itudinal

-0.58

Rub

-0.57

 recapt

-0.55

âĹ¼

-0.54

ivals

-0.54

worthiness

-0.54

eor

-0.53

åħ

-0.52

POSITIVE LOGITS

 opposed

1.14

pired

1.11

pires

0.99

bestos

0.97

cription

0.93

 well

0.92

phy

0.92

well

0.90

part

0.89

ynchron

0.89

Activations Density 0.182%

comparative phrases, particularly using the word "as" followed by an adjective or adverb

phrases indicating comparison or similarity

No Comments

No Known Activations