INDEX

Explanations

terms related to directing or giving instructions

oai_token-act-pair · gpt-3.5-turbo

expressively articulated desires or wishes

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

 agre

-0.81

 cris

-0.73

 trump

-0.70

 overpower

-0.70

 destro

-0.70

 sculpt

-0.69

 streng

-0.69

ĸļ

-0.69

oun

-0.68

 thresholds

-0.68

POSITIVE LOGITS

Reply

1.18

Topics

1.16

////////////////////////////////

1.12

Posts

1.11

--------------------------------------------------------

1.09

Advertisements

1.07

--------------------------------

1.02

--------------------

1.01

Translation

1.01

Pand

1.00

Activations Density 0.367%

terms related to directing or giving instructions

expressively articulated desires or wishes

No Comments

No Known Activations