INDEX

Explanations

references to information mentioned or said earlier in a text

oai_token-act-pair · gpt-3.5-turbo

phrases that refer to prior statements or mentions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

istries

-0.85

fu

-0.80

Made

-0.79

mens

-0.71

istered

-0.70

bled

-0.69

mop

-0.67

wear

-0.67

ores

-0.67

stru

-0.66

POSITIVE LOGITS

 earlier

1.18

 previously

0.94

 above

0.88

 previous

0.75

 deduction

0.73

 acknow

0.73

 aforementioned

0.72

mentioned

0.72

 yesterday

0.70

 hitherto

0.70

Activations Density 0.087%

references to information mentioned or said earlier in a text

phrases that refer to prior statements or mentions

No Comments

No Known Activations