INDEX

Explanations

verbs related to describing or explaining something

oai_token-act-pair · gpt-3.5-turbo

terms associated with description, commemoration, and representation

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uden

-0.67

luaj

-0.66

udi

-0.64

authorized

-0.63

edin

-0.62

iked

-0.59

ãģ®

-0.59

appointed

-0.58

numbered

-0.58

POSITIVE LOGITS

uate

0.72

 conve

0.65

 differentiation

0.63

 subsistence

0.62

BG

0.62

 gradient

0.61

itial

0.61

 optimum

0.61

 atmospheric

0.60

 overflow

0.60

Activations Density 0.308%

verbs related to describing or explaining something

terms associated with description, commemoration, and representation

No Comments

No Known Activations