INDEX

Explanations

words related to diversity and variety, particularly in contexts like social settings and objects

oai_token-act-pair · gpt-3.5-turbo

terms related to qualities of being diverse or fragmented

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

rir

-0.71

orders

-0.69

 drilled

-0.67

Tor

-0.61

erve

-0.61

raz

-0.61

href

-0.60

flag

-0.59

Tube

-0.58

srf

-0.58

POSITIVE LOGITS

enance

0.77

theless

0.70

glers

0.68

 minded

0.67

nesses

0.66

tenance

0.64

ly

0.64

(<

0.61

ishly

0.61

heid

0.60

Activations Density 0.142%

words related to diversity and variety, particularly in contexts like social settings and objects

terms related to qualities of being diverse or fragmented

No Comments

No Known Activations