INDEX

Explanations

adjectives describing extreme conditions or situations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "severe" and associated severity-related terms

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

xxxxxxxx

-0.79

ynthesis

-0.78

atu

-0.76

rium

-0.75

cellent

-0.75

 Tycoon

-0.74

apolis

-0.72

aukee

-0.71

wagen

-0.70

etts

-0.70

POSITIVE LOGITS

 punishments

0.90

 severity

0.85

 punishment

0.85

 enough

0.83

 depressive

0.83

 earthqu

0.82

 repercussions

0.82

 consequences

0.81

 understatement

0.81

 traumatic

0.81

Activations Density 0.020%

adjectives describing extreme conditions or situations

instances of the word "severe" and associated severity-related terms

No Comments

No Known Activations