INDEX

Explanations

plausible explanations or reasons

oai_token-act-pair · gpt-3.5-turbo

variations of the word "plausible" and its related forms

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

sterdam

-0.74

ounty

-0.74

angelo

-0.74

usterity

-0.73

enfranch

-0.73

elight

-0.72

zona

-0.72

une

-0.71

eg

-0.70

cellence

-0.69

POSITIVE LOGITS

ausible

0.89

\\\\\\\\

0.86

 explanation

0.82

 plausible

0.82

 excuse

0.81

 explanations

0.80

 plaus

0.80

den

0.78

 guesses

0.76

 scenario

0.76

Activations Density 0.028%

plausible explanations or reasons

variations of the word "plausible" and its related forms

No Comments

No Known Activations