INDEX

Explanations

adjectives that may be related to evaluation or classification

oai_token-act-pair · gpt-3.5-turbo

words related to academic achievement and success

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

 Lines

-0.62

 Accountability

-0.60

 Balanced

-0.59

 Izan

-0.57

 Commons

-0.57

 Pyramid

-0.55

 Empires

-0.54

 Shuttle

-0.53

Own

-0.53

 Borders

-0.53

POSITIVE LOGITS

ole

0.64

aging

0.60

ect

0.60

ote

0.59

atur

0.59

Äĵ

0.58

ouse

0.58

asing

0.58

aded

0.58

ÂŃ

0.58

Activations Density 0.372%

adjectives that may be related to evaluation or classification

words related to academic achievement and success

No Comments

No Known Activations