INDEX

Explanations

the word "this" followed by further information or context

oai_token-act-pair · gpt-3.5-turbo

instances of the word "this" indicating focus on specific events or subjects

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

ãĤ¹ãĥĪ

-0.81

ãĥ¼ãĥ³

-0.80

istries

-0.75

ãĥ©ãĥ³

-0.75

anamo

-0.73

isms

-0.72

adle

-0.72

ãĤ¬

-0.72

cks

-0.71

ortment

-0.71

POSITIVE LOGITS

 trope

0.94

 particular

0.90

 arrangement

0.90

 happened

0.84

 newfound

0.83

 phenomenon

0.83

 enigmatic

0.82

 nifty

0.81

 latest

0.80

guy

0.80

Activations Density 0.139%

the word "this" followed by further information or context

instances of the word "this" indicating focus on specific events or subjects

No Comments

No Known Activations