INDEX

Explanations

ambiguous or uncertain language

oai_token-act-pair · gpt-3.5-turbo

expressions of uncertainty or lack of clarity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

azar

-0.79

arms

-0.77

rather

-0.76

zbollah

-0.74

 ACTIONS

-0.71

paio

-0.67

MAY

-0.65

oubted

-0.65

YES

-0.65

gdala

-0.63

POSITIVE LOGITS

 anymore

1.66

nor

1.30

yet

1.03

yet

0.96

 necessarily

0.92

 anything

0.88

 anybody

0.86

 whatsoever

0.86

any

0.83

 anywhere

0.82

Activations Density 0.320%

ambiguous or uncertain language

expressions of uncertainty or lack of clarity

No Comments

No Known Activations