INDEX

Explanations

phrases related to clapping or expressing approval through physical gestures

oai_token-act-pair · gpt-3.5-turbo

words relating to the concept of being constrained or limited

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

 toll

-0.73

®

-0.68

neg

-0.68

gio

-0.64

FY

-0.62

 athe

-0.61

cig

-0.61

 Fenrir

-0.60

©¶æ

-0.60

gary

-0.60

POSITIVE LOGITS

oleon

0.91

urated

0.85

olean

0.85

ropri

0.84

itone

0.82

 GOODMAN

0.80

olitan

0.79

*/(

0.78

ocalypse

0.78

olicy

0.75

Activations Density 0.018%

phrases related to clapping or expressing approval through physical gestures

words relating to the concept of being constrained or limited

No Comments

No Known Activations