INDEX

Explanations

phrases related to conditions or actions being provided or given

oai_token-act-pair · gpt-3.5-turbo

the word "provided" in various contexts

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

focus

-0.76

MAT

-0.74

phase

-0.71

can

-0.70

Rated

-0.69

spect

-0.68

pred

-0.68

arios

-0.67

do

-0.66

Boss

-0.66

POSITIVE LOGITS

utical

0.90

 sust

0.74

atives

0.73

 refunds

0.73

 supplied

0.71

 assistance

0.71

Ö¼

0.70

amera

0.70

ordial

0.69

 condolences

0.69

Activations Density 0.027%

phrases related to conditions or actions being provided or given

the word "provided" in various contexts

No Known Activations