INDEX

Explanations

instances of the word "say" followed by a rating number

oai_token-act-pair · gpt-3.5-turbo

expressions and phrases related to stating opinions or declarations

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

xtap

-0.84

ntil

-0.81

phrine

-0.80

ritional

-0.79

¥ŀ

-0.76

 cumbers

-0.74

mental

-0.73

folios

-0.72

pleting

-0.71

ptives

-0.70

POSITIVE LOGITS

 goodbye

1.48

 hello

1.00

 aloud

0.94

 farewell

0.84

 Goodbye

0.82

 loudly

0.77

bye

0.77

ieu

0.77

hi

0.74

hey

0.68

Activations Density 0.134%

instances of the word "say" followed by a rating number

expressions and phrases related to stating opinions or declarations

No Comments

No Known Activations