INDEX

Explanations

phrases related to statements or verbal communication

oai_token-act-pair · gpt-3.5-turbo

instances of the verb "say" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ngth

-0.80

ãĤ¼

-0.78

ãĤ´ãĥ³

-0.64

Ĥª

-0.64

 Nanto

-0.64

Ö¼

-0.63

âĹ¼

-0.61

ptic

-0.60

ãĥ¯

-0.59

Tur

-0.57

POSITIVE LOGITS

 definitively

1.31

 anything

1.30

 whether

1.22

 aloud

1.06

 unequivocally

1.06

anything

1.04

 exactly

1.03

 goodbye

1.00

 explicitly

0.99

why

0.96

Activations Density 0.052%

phrases related to statements or verbal communication

instances of the verb "say" and its variations

No Comments

No Known Activations