INDEX

Explanations

adjectives and verbs related to personal experiences and emotions

oai_token-act-pair · gpt-3.5-turbo

expressions of emotional states or feelings

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

 policymakers

-0.66

 multinational

-0.63

 internationally

-0.61

ortium

-0.61

 Infrastructure

-0.60

 NATO

-0.59

 proponents

-0.59

 incumbent

-0.59

 nationally

-0.58

 Marketplace

-0.58

POSITIVE LOGITS

 haha

0.77

;)

0.71

 fuckin

0.67

atos

0.66

Laughs

0.65

tin

0.63

-"

0.63

:)

0.62

lol

0.62

eeee

0.61

Activations Density 1.912%

adjectives and verbs related to personal experiences and emotions

expressions of emotional states or feelings

No Comments

No Known Activations