INDEX

Explanations

references to reactions or replies

oai_token-act-pair · gpt-3.5-turbo

mentions of varying responses or reactions

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

rome

-0.81

cutting

-0.72

ramer

-0.69

ffe

-0.67

hemat

-0.67

rafted

-0.64

rust

-0.63

rip

-0.62

raft

-0.62

roots

-0.62

POSITIVE LOGITS

 thereto

0.97

 response

0.94

 responses

0.90

 reaction

0.82

ivation

0.79

ively

0.79

naires

0.78

DragonMagazine

0.77

response

0.77

 reactions

0.77

Activations Density 0.031%

references to reactions or replies

mentions of varying responses or reactions

No Comments

No Known Activations