INDEX

Explanations

phrases signaling a change in topic or perspective

oai_token-act-pair · gpt-3.5-turbo

repeated conjunctions that indicate a list or accumulation of thoughts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ãģĨ

-0.70

ABE

-0.64

pees

-0.64

ses

-0.64

sense

-0.64

HB

-0.63

hoe

-0.62

sic

-0.62

cloth

-0.61

xes

-0.60

POSITIVE LOGITS

romeda

1.66

rea

1.37

ersen

1.04

hra

1.03

rew

0.93

 secondly

0.91

uin

0.90

rogens

0.89

alus

0.87

rey

0.82

Activations Density 0.056%

phrases signaling a change in topic or perspective

repeated conjunctions that indicate a list or accumulation of thoughts

No Comments

No Known Activations