INDEX

Explanations

phrases related to assessing a situation and taking appropriate action

oai_token-act-pair · gpt-3.5-turbo

the word "accordingly" in various contexts related to responses or actions taken based on situations or conditions

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

IFrame

Link

Negative Logits

 Franks

-0.84

ker

-0.72

 Roses

-0.71

 Mehran

-0.70

oppable

-0.69

cock

-0.67

 Feld

-0.66

 Lyme

-0.65

dan

-0.65

 Fulton

-0.63

POSITIVE LOGITS

 accordingly

0.80

ãĤ¼ãĤ¦ãĤ¹

0.79

itiz

0.74

 dilig

0.74

é»Ĵ

0.74

peror

0.72

é¾įåĸļå£«

0.72

 citiz

0.71

 behaved

0.71

edi

0.71

Activations Density 0.011%

phrases related to assessing a situation and taking appropriate action

the word "accordingly" in various contexts related to responses or actions taken based on situations or conditions

No Known Activations