INDEX

Explanations

phrases related to negative impacts or destruction

oai_token-act-pair · gpt-3.5-turbo

instances of the word "the" in various contexts

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

 wisely

-0.72

 namely

-0.71

Ã©e

-0.68

 alone

-0.67

rand

-0.63

 lets

-0.63

 signifies

-0.62

icho

-0.61

ESPN

-0.61

ctl

-0.60

POSITIVE LOGITS

 entire

1.42

 entirety

1.23

 remainder

1.16

 whole

1.13

 slightest

1.07

 latter

1.06

 possibility

1.04

 same

1.00

 proverbial

0.97

 notion

0.97

Activations Density 0.462%

phrases related to negative impacts or destruction

instances of the word "the" in various contexts

No Comments

No Known Activations