INDEX

Explanations

phrases related to achieving success or completion

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "break through"

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

ials

-0.74

ise

-0.71

arily

-0.71

juven

-0.71

Ring

-0.70

ises

-0.70

izes

-0.69

rouse

-0.64

abet

-0.63

isd

-0.62

POSITIVE LOGITS

 waivers

0.75

oresc

0.70

agall

0.68

 Rodgers

0.68

erella

0.65

 customs

0.64

lace

0.64

ework

0.63

heit

0.62

 intermedi

0.61

Activations Density 0.040%

phrases related to achieving success or completion

instances of the phrase "break through"

No Comments

No Known Activations