INDEX

Explanations

questions ending in '?'

oai_token-act-pair · gpt-3.5-turbo

sentences that contain questions or exclamations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

amas

-0.64

duct

-0.62

heastern

-0.61

isol

-0.61

ciplinary

-0.61

hran

-0.60

anos

-0.59

oreal

-0.58

 reven

-0.58

onis

-0.57

POSITIVE LOGITS

 Then

1.20

 Well

1.08

 Nope

1.06

 Immediately

1.04

 Later

1.01

And

1.00

 Because

0.99

 Which

0.98

 Yeah

0.97

 Suddenly

0.95

Activations Density 0.143%

questions ending in '?'

sentences that contain questions or exclamations

No Comments

No Known Activations