INDEX

Explanations

mentions of geographical locations, specifically related to bodies of water

oai_token-act-pair · gpt-3.5-turbo

vocabulary related to marine or aquatic environments

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

 Bened

-0.73

ADRA

-0.70

Interstitial

-0.69

catentry

-0.67

 objections

-0.65

 Palm

-0.59

 introductory

-0.59

Lif

-0.59

 Bravo

-0.58

 Garc

-0.58

POSITIVE LOGITS

olitics

0.81

olit

0.80

oard

0.79

inis

0.77

ird

0.75

etch

0.73

ind

0.73

opolis

0.71

illance

0.71

reen

0.71

Activations Density 0.099%

mentions of geographical locations, specifically related to bodies of water

vocabulary related to marine or aquatic environments

No Comments

No Known Activations