INDEX

Explanations

mentions of potential actions or tasks

oai_token-act-pair · gpt-3.5-turbo

variations in numerical values or monetary concepts

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

 emancipation

-0.81

 scripting

-0.77

 treaty

-0.74

 frontline

-0.73

 treaties

-0.71

 goalkeeper

-0.71

 derby

-0.70

 tram

-0.70

 resolve

-0.70

 quir

-0.70

POSITIVE LOGITS

Advertisements

1.22

Interstitial

1.09

If

1.08

https

1.07

Feature

1.07

"âĢ¦

1.07

Tumblr

1.06

âĢ¢

1.06

"(

1.06

Why

1.05

Activations Density 0.126%

mentions of potential actions or tasks

variations in numerical values or monetary concepts

No Comments

No Known Activations