INDEX

Explanations

phrases comparing or contrasting different entities or concepts

oai_token-act-pair · gpt-3.5-turbo

the phrase "of" in various contexts, often indicating comparison or relation

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

 showc

-0.77

 ende

-0.74

 traged

-0.69

 pestic

-0.69

 cryst

-0.69

 reperto

-0.68

 livest

-0.68

 misunder

-0.68

isively

-0.65

 revol

-0.64

POSITIVE LOGITS

icial

0.82

mund

0.74

 course

0.71

tains

0.69

 ordinary

0.68

ramer

0.66

course

0.64

 Brill

0.63

Ħ

0.63

¢

0.63

Activations Density 0.054%

phrases comparing or contrasting different entities or concepts

the phrase "of" in various contexts, often indicating comparison or relation

No Comments

No Known Activations