INDEX

Explanations

questions and statements related to analysis and comparison

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate uncertainty or complexity in arguments

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

SPONSORED

-0.87

agara

-0.61

".

-0.61

âĢ¦."

-0.61

 guiActiveUnfocused

-0.60

"},{"

-0.59

"}],"

-0.58

javascript

-0.58

)",

-0.56

EStream

-0.56

POSITIVE LOGITS

oret

0.70

cknowled

0.58

 Firstly

0.57

itialized

0.54

sofar

0.54

Especially

0.53

neath

0.51

 consequence

0.49

 Wand

0.49

Firstly

0.49

Activations Density 0.788%

questions and statements related to analysis and comparison

phrases that indicate uncertainty or complexity in arguments

No Comments

No Known Activations