INDEX

Explanations

medical conditions or health-related terms

oai_token-act-pair · gpt-3.5-turbo

instances of the verb "is" 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ends

-0.74

iates

-0.70

perate

-0.68

agues

-0.67

iate

-0.66

 ends

-0.65

eor

-0.64

essor

-0.62

urches

-0.61

uum

-0.61

POSITIVE LOGITS

 undoubtedly

0.92

ometric

0.83

 doubtless

0.72

ovie

0.70

cussion

0.69

 indeed

0.68

 probably

0.68

 evidenced

0.68

 termed

0.67

omorphic

0.66

Activations Density 0.103%

medical conditions or health-related terms

instances of the verb "is" in various contexts

No Comments

No Known Activations