INDEX

Explanations

phrases related to academic institutions or research organizations

oai_token-act-pair · gpt-3.5-turbo

references to various institutes or organizations

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

theless

-1.01

 tame

-0.71

 ages

-0.71

 sled

-0.70

mere

-0.68

seys

-0.68

taboola

-0.67

nir

-0.67

=-=-=-=-=-=-=-=-

-0.67

catentry

-0.66

POSITIVE LOGITS

 Transparency

0.78

 Square

0.78

una

0.77

itute

0.77

Research

0.71

 Research

0.71

 Institute

0.70

fusc

0.70

 Institution

0.69

 Herz

0.68

Activations Density 0.020%

phrases related to academic institutions or research organizations

references to various institutes or organizations

No Comments

No Known Activations