INDEX

Explanations

phrases related to experts and analysts in various fields

oai_token-act-pair · gpt-3.5-turbo

references to various types of experts and their opinions in the text

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

Enough

-0.67

Interstitial

-0.67

Ore

-0.65

comes

-0.65

Dialogue

-0.62

 Horus

-0.61

Quantity

-0.61

 Instruction

-0.60

HAHA

-0.60

Complete

-0.59

POSITIVE LOGITS

 speculate

1.39

 contend

1.29

say

1.21

 argue

1.20

 warn

1.19

 believe

1.19

 disagree

1.17

 suggest

1.16

 predict

1.15

 anticipate

1.14

Activations Density 0.131%

phrases related to experts and analysts in various fields

references to various types of experts and their opinions in the text

No Comments

No Known Activations