INDEX

Explanations

personal pronouns followed by a negative and the verb "t" (e.g. don't, can't, won't)

oai_token-act-pair · gpt-3.5-turbo

the repeated phrase "didn't" 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 @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

inia

-0.76

 Reduced

-0.67

 Groups

-0.66

 Suff

-0.64

 Compar

-0.58

 Medium

-0.58

å£«

-0.58

 arsen

-0.57

lain

-0.57

 Alleg

-0.57

POSITIVE LOGITS

 bother

1.02

 realise

0.93

 hesitate

0.90

 realize

0.90

 dare

0.87

erest

0.86

want

0.83

 survive

0.80

 necessarily

0.80

 comply

0.78

Activations Density 0.081%

personal pronouns followed by a negative and the verb "t" (e.g. don't, can't, won't)

the repeated phrase "didn't" in various contexts

No Comments

No Known Activations