INDEX

Explanations

signs or indications of various contexts or situations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating signs of progress or change

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

 Arena

-0.67

 pores

-0.64

 cabinets

-0.64

boro

-0.64

ken

-0.62

rients

-0.62

uge

-0.61

ggle

-0.61

FFER

-0.61

bers

-0.60

POSITIVE LOGITS

 customary

0.72

 precaution

0.71

 tradition

0.70

[|

0.64

 derog

0.64

onymous

0.62

 Cyrus

0.61

 Shutdown

0.60

 Hipp

0.59

 srfAttach

0.59

Activations Density 0.504%

signs or indications of various contexts or situations

phrases indicating signs of progress or change

No Comments

No Known Activations