INDEX

Explanations

terms related to intense or fervent debates, conflicts, or situations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "heated" in various contexts

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

ogo

-0.86

roma

-0.78

ateral

-0.78

ystem

-0.75

ment

-0.75

ees

-0.73

pha

-0.68

ledged

-0.68

oro

-0.67

itte

-0.64

POSITIVE LOGITS

 heated

1.08

 heating

0.92

 uproar

0.84

ãĤ¦ãĤ¹

0.84

rawdownloadcloneembedreportprint

0.78

ly

0.78

 heats

0.76

 overhe

0.75

 temperatures

0.73

 currents

0.68

Activations Density 0.033%

terms related to intense or fervent debates, conflicts, or situations

occurrences of the word "heated" in various contexts

No Comments

No Known Activations