INDEX

Explanations

mentions of actions or events related to people and significant situations, possibly in a journalistic context

oai_token-act-pair · gpt-3.5-turbo

key actions or themes related to storytelling in various media

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

thin

-0.66

factor

-0.64

 syndrome

-0.64

soType

-0.61

itized

-0.59

itu

-0.59

lear

-0.57

pee

-0.57

ween

-0.56

Hat

-0.55

POSITIVE LOGITS

unda

0.66

 mainly

0.61

 essentially

0.61

 acronym

0.61

 roughly

0.60

actionDate

0.59

 approximately

0.59

atform

0.59

 broadly

0.57

yrinth

0.57

Activations Density 0.519%

mentions of actions or events related to people and significant situations, possibly in a journalistic context

key actions or themes related to storytelling in various media

No Comments

No Known Activations