INDEX

Explanations

phrases related to technical instructions or descriptions

oai_token-act-pair · gpt-3.5-turbo

the word "as" in various contexts

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

ople

-0.67

Rub

-0.63

ATIVE

-0.63

rous

-0.62

quist

-0.62

rolet

-0.61

ries

-0.59

osi

-0.59

Posts

-0.58

OY

-0.58

POSITIVE LOGITS

ynchron

1.18

pired

1.08

 well

1.06

 follows

1.06

 opposed

1.06

pires

1.03

phy

1.03

cription

0.97

well

0.96

bestos

0.96

Activations Density 0.229%

phrases related to technical instructions or descriptions

the word "as" in various contexts

No Comments

No Known Activations